Merge remote-tracking branch 'origin/feat/directory-picker-quiet-navigation' into feat/dir-selector-adaptive-default

# Conflicts:
#	.agents/notes/implemented/architecture/2026-07-28-directory-picker-capability-seam.i18n.yaml
This commit is contained in:
creatixchu
2026-07-30 16:20:22 +08:00
95 changed files with 5602 additions and 632 deletions

View File

@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write packages/core/system-prompt/README.md
README.md: 79badba0b84b27c01f25e9c31b5df78c556411ea
README.zh.md: fb3ed08bc9ea546033acac3b577980f500ce243d
README.md: 23bc0e8177ad2a778df9522e254bfd5e03a9871f
README.zh.md: 1fd4febc1c15acda19e7abfca94079b9585c1972

View File

@@ -8,6 +8,7 @@ System prompt assembly registry. Plugins contribute ordered sections, tool schem
| Key | Default | Meaning |
|---|---|---|
| `includeHarnessIdentity` | `true` | Include the fixed `You are an AI agent powered by the DeepSeek Harness SDK.` order-100 opener. Set false only when a compatibility deployment owns the complete system prompt. |
| `persona` | `''` | The global deployment-persona default: the ONE config-authored prompt fragment, rendered as the order-0 `deployment:persona` section unless an agent-scoped contribution shadows it. A template — complete `{{…}}` groups are interpreted strictly against the registered variables (the shipped loop registers `{{model}}`/`{{cwd}}`), with no escape syntax for literal braces yet. Empty ⇒ the section is dropped at render. |
| `toolOrder` | — | Explicit model-facing tool order, as a list of `ToolSchema.name`s with one `'<unlisted-tools>'` rest entry (`TOOL_ORDER_REST`): listed tools take their listed position, unlisted tools land at the rest entry in lexicographic name order. Absent ⇒ plain lexicographic name order. Applied to the collected tools BEFORE the `system-prompt/assemble` waterfall — like the sections' `order` sort, it canonicalizes what the registry contributed (registration order is a plugin-load artifact), and a waterfall listener that mutates the list owns the determinism of what it emits. Misconfiguration fails loud: a list without exactly one rest entry, or with duplicates, throws at load; a listed name with no registered tool rejects every `assemble()`; a tool provider returning the reserved rest-entry name also rejects. Under the shipped loop the turn fails before any model request. Why a central list and not per-plugin weights: [Explicit model-facing tool order](../../../.agents/notes/implemented/feature/2026-07-06-explicit-tool-order.md). |
@@ -48,7 +49,7 @@ Design rationale: [the prompt-variables Agent Note](../../../.agents/notes/imple
#### What the model sees
Every assembly starts with the harness identity below, then the configured persona and ordered plugin sections after strict variable interpolation. Empty sections disappear; scoped sections and variables can shadow globals for one agent. The final `system-prompt/assemble` waterfall result is authoritative, so an expert listener's changes determine the delivered prompt and tool schemas.
By default every assembly starts with the harness identity below, then the configured persona and ordered plugin sections after strict variable interpolation. `includeHarnessIdentity: false` omits only that fixed opener for a deployment that owns the complete compatibility persona. Empty sections disappear; scoped sections and variables can shadow globals for one agent. The final `system-prompt/assemble` waterfall result is authoritative, so an expert listener's changes determine the delivered prompt and tool schemas.
##### Harness identity
@@ -58,7 +59,7 @@ You are an AI agent powered by the DeepSeek Harness SDK.
#### Token effect
Identity is a fixed per-request cost. Persona and plugin text are repeated per request and scale with their rendered content.
Identity is a fixed per-request cost when enabled. Persona and plugin text are repeated per request and scale with their rendered content.
#### KV Cache effect

View File

@@ -8,6 +8,7 @@
| 键 | 默认值 | 含义 |
|---|---|---|
| `includeHarnessIdentity` | `true` | 是否包含固定的 `You are an AI agent powered by the DeepSeek Harness SDK.`、顺序为 100 的开场白。仅当兼容部署拥有完整系统提示词时设为 false。 |
| `persona` | `''` | 全局部署 persona 默认值:唯一由配置创作的提示词片段,渲染为顺序为 0 的 `deployment:persona` 段,除非 agent 作用域的贡献将其遮蔽。它是模板,完整的 `{{…}}` 组会严格按已注册变量解释(随附循环注册 `{{model}}`/`{{cwd}}`),目前没有表达字面量花括号的转义语法。为空 ⇒ 渲染时删除该段。 |
| `toolOrder` | 无 | 显式的面向模型工具顺序:一个 `ToolSchema.name` 列表,包含一个 `'<unlisted-tools>'` 其余项(`TOOL_ORDER_REST`)。已列工具占据列出的位置;未列工具按名称字典序落在其余项位置。缺席 ⇒ 直接按名称字典序排列。在 `system-prompt/assemble` waterfall瀑布式事件之前应用于已收集工具与段的 `order` 排序一样,它会规范化注册表贡献的内容(注册顺序是插件加载产物),而修改列表的 waterfall 监听器拥有其输出的确定性。配置错误会明确失败:列表没有恰好一个其余项或存在重复项,会在加载时抛出;已列名称没有对应已注册工具,会使每次 `assemble()` 被拒绝;工具提供方返回保留的其余项名称也会被拒绝。在随附循环下,轮次会在任何模型请求前失败。为何采用中心列表而非每插件权重,见[显式面向模型工具顺序](../../../.agents/notes/implemented/feature/2026-07-06-explicit-tool-order.md)。 |
@@ -48,7 +49,7 @@
#### 模型看到的内容
每次组装都从下方 harness 身份开始,然后在严格变量插值后追加已配置 persona 与有序插件段。空段会消失;带作用域的段和变量可以为一个 agent 遮蔽全局项。最终 `system-prompt/assemble` waterfall 结果是权威来源,因此专家监听器的变更决定交付的提示词与工具 schema。
默认情况下,每次组装都从下方 harness 身份开始,然后在严格变量插值后追加已配置 persona 与有序插件段。`includeHarnessIdentity: false` 仅为拥有完整兼容 persona 的部署省略这个固定开场白。空段会消失;带作用域的段和变量可以为一个 agent 遮蔽全局项。最终 `system-prompt/assemble` waterfall 结果是权威来源,因此专家监听器的变更决定交付的提示词与工具 schema。
##### Harness 身份
@@ -58,7 +59,7 @@ You are an AI agent powered by the DeepSeek Harness SDK.
#### Token 影响
身份是每次请求的固定成本。Persona 与插件文本在每次请求中重复,成本随渲染内容增长。
启用时,身份是每次请求的固定成本。Persona 与插件文本在每次请求中重复,成本随渲染内容增长。
#### KV Cache 影响

View File

@@ -145,6 +145,8 @@ function compareToolNames(a: ToolSchema, b: ToolSchema): number {
/** Plugin config: the deployment-authored fragment of the system prompt (see {@link Config.persona} for its contract). */
export interface Config {
/** Include the fixed DeepSeek Harness identity before the deployment persona (default true). */
includeHarnessIdentity?: boolean
/**
* Deployment-wide order-0 persona template. A scoped section named
* `deployment:persona` shadows it; `{{variable}}` references are strict.
@@ -245,6 +247,7 @@ class PromptLayer implements ScopeLayer {
/** Registry service for the prompt inputs assembled before each model step. */
export class SystemPrompt extends Service {
static Config: z<Config> = z.object({
includeHarnessIdentity: z.boolean().default(true),
persona: z.string().default(''),
// Preserve omission because an explicit empty order lacks the rest marker.
toolOrder: z.array(z.string()).default(undefined as unknown as string[]),
@@ -260,11 +263,13 @@ export class SystemPrompt extends Service {
super(ctx, 'systemPrompt')
this.toolOrder = validateToolOrder(config.toolOrder)
// Keep harness-owned openers independent of the selected loop plugin.
this.section({
name: 'harness:identity',
order: -100,
text: 'You are an AI agent powered by the DeepSeek Harness SDK.',
})
if (config.includeHarnessIdentity ?? true) {
this.section({
name: 'harness:identity',
order: -100,
text: 'You are an AI agent powered by the DeepSeek Harness SDK.',
})
}
this.section({
name: 'deployment:persona',
order: 0,

View File

@@ -37,6 +37,18 @@ describe('SystemPrompt', () => {
expect(renderPrompt(await ctx.systemPrompt.assemble())).toBe(IDENTITY)
})
it('can omit the harness identity for a deployment that owns the complete persona', async () => {
const ctx = new Context()
await ctx.plugin(SystemPrompt, {
includeHarnessIdentity: false,
persona: 'You are a helpful software engineer assistant.',
})
const assembly = await ctx.systemPrompt.assemble()
expect(assembly.sections.map(section => section.name)).toEqual(['deployment:persona'])
expect(renderPrompt(assembly)).toBe('You are a helpful software engineer assistant.')
})
it('tolerates a schema-bypassing direct construction (persona omitted)', async () => {
// ctx.plugin validates + defaults the config first; a direct construction
// skips the schema, so the ctor's `?? ''` narrowing is what fires.

View File

@@ -23,7 +23,7 @@ describe('gen-tool-catalog collectToolCatalog', () => {
it('boots every shipped tool package and harvests its model-facing schemas', async () => {
const catalog = await collectToolCatalog()
const names = catalog.flatMap(entry => entry.schemas.map(s => s.name)).sort()
expect(names).toEqual(['ask_user_question', 'bash', 'cordis_inspect', 'cordis_mount', 'cordis_unmount', 'create_goal', 'edit', 'exit_plan_mode', 'get_goal', 'glob', 'grep', 'lsp', 'ralph', 'read', 'run_code', 'session_event_read', 'session_event_search', 'session_event_trace', 'session_search', 'session_trace', 'skill', 'subagent', 'task_kill', 'task_list', 'task_output', 'terminal_close', 'terminal_list', 'terminal_open', 'terminal_read', 'terminal_send', 'terminal_signal', 'todo_write', 'update_goal', 'web_fetch', 'web_search', 'workflow', 'write'])
expect(names).toEqual(['ask_user_question', 'bash', 'bash', 'cordis_inspect', 'cordis_mount', 'cordis_unmount', 'create_goal', 'edit', 'exit_plan_mode', 'get_goal', 'glob', 'grep', 'lsp', 'ralph', 'read', 'run_code', 'session_event_read', 'session_event_search', 'session_event_trace', 'session_search', 'session_trace', 'skill', 'str_replace_editor', 'subagent', 'task_kill', 'task_list', 'task_output', 'terminal_close', 'terminal_list', 'terminal_open', 'terminal_read', 'terminal_send', 'terminal_signal', 'todo_write', 'update_goal', 'web_fetch', 'web_search', 'workflow', 'write'])
// Every tool carries a JSON-Schema `parameters` object (what the model sees).
for (const entry of catalog) {
for (const schema of entry.schemas) {

View File

@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write packages/examples/agent-spine-demo/README.md
README.md: 359e7153be2f480ba3fea4b06782acdc9f89ebb9
README.zh.md: acd8c06940b03e90e368314cd725846a2b92b656
README.md: 6bbd99217bcdce0e8a0e8fd22a8d39d0c64224b9
README.zh.md: 4a7e7a1b7eced9317f94eeddf2442ded9a3e0e13

View File

@@ -31,7 +31,7 @@ Read this package for the whole plugin tree and its composition order.
@deepseek-ai/dsh-scope/invariant
@deepseek-ai/dsh-agent-loop/invariant
package-owned relational checks
@deepseek-ai/dsh-tool-bash the model-facing bash schema
@deepseek-ai/dsh-tool-bash the model-facing bash schema (unless toolBash=false)
@deepseek-ai/dsh-workspace-context AGENTS.md/CLAUDE.md workspace context loader
@deepseek-ai/dsh-tool-skill session-prefix skill catalog + model-facing loader schema
@deepseek-ai/dsh-tool-tasks task_output/task_list/task_kill schemas + completion notices
@@ -55,11 +55,11 @@ This is the [interface/implementation/consumer seam](../../../.agents/notes/impl
```ts
import type { Config } from '@deepseek-ai/dsh-agent-spine-demo'
// { agents?, maxParallelToolCalls?, persona?, toolOrder?, tools?, dshHome?, sessionTitle?, skills?, workspaceContext, toolBash?, toolTasks?, goals?, invariants? }
// { agents?, maxParallelToolCalls?, includeHarnessIdentity?, persona?, toolOrder?, tools?, dshHome?, sessionTitle?, skills?, workspaceContext, toolBash?, toolTasks?, goals?, invariants? }
// workspaceContext requires { maxBytes } or false; the other owner schemas supply defaults.
```
The bundle FORWARDS each field to the child that owns it: `agents` and `maxParallelToolCalls` to `agent-loop` (`agents` defaults to `[]`; the cap defaults there), so each app supplies its own pre-created agents — TUI and headless apps pre-create `main`, while the ACP app creates agents on demand at `session/new`; `persona` and `toolOrder` to `dsh-system-prompt`; `tools` to the tool registry for its presentation mode; `sessionTitle` to the fallback title service; `skills.registry`, `skills.local`, and `skills.tool` to the skill registry, local provider, and model-facing consumer; the required `workspaceContext` choice to `dsh-workspace-context` (`{ maxBytes }` enables loading and `false` disables it); `invariants` to the invariant service; and `toolBash`/`toolTasks` to the two model-facing tool plugins the bundle owns. It always mounts `dsh-llm-retry`, while each leaf adapter owns its nested `retryPolicy`. Omitted `sessionTitle` uses the explicit example policy of 5 words, 40 fallback bytes, and 80 accepted-title bytes. A `goals` object opts into the persisted domain, model tools, and same-session driver while forwarding `goals.domain` and `goals.tool` to their owners; omission or `false` leaves the stack absent so headless callers retain one-turn settlement. Set `skills.enabled: false` to omit both the local provider and model-facing skill tool, and set `toolTasks: false` to retain the task service for foreground producers without exposing `task_output` / `task_list` / `task_kill`. It resolves `dshHome` once through [`@deepseek-ai/dsh-paths`](../../util/paths/README.md) and forwards that absolute value to tool-bash's managed environment and enabled local skill discovery. An absent top-level `dshHome` adopts `skills.local.dshHome`; supplying both with different resolved paths fails loudly. `toolBash.enableRunInBackground` controls only the bash producer; independently loaded producers keep their own config. Workspace instructions register before the skill catalog so their session-prefix message renders first. App packages use `pickSpineConfig()` to copy only these bundle-owned fields.
The bundle FORWARDS each field to the child that owns it: `agents` and `maxParallelToolCalls` to `agent-loop` (`agents` defaults to `[]`; the cap defaults there), so each app supplies its own pre-created agents — TUI and headless apps pre-create `main`, while the ACP app creates agents on demand at `session/new`; `includeHarnessIdentity`, `persona`, and `toolOrder` to `dsh-system-prompt`; `tools` to the tool registry for its presentation mode; `sessionTitle` to the fallback title service; `skills.registry`, `skills.local`, and `skills.tool` to the skill registry, local provider, and model-facing consumer; the required `workspaceContext` choice to `dsh-workspace-context` (`{ maxBytes }` enables loading and `false` disables it); `invariants` to the invariant service; and `toolBash`/`toolTasks` to the two model-facing tool plugins the bundle owns. It always mounts `dsh-llm-retry`, while each leaf adapter owns its nested `retryPolicy`. Omitted `sessionTitle` uses the explicit example policy of 5 words, 40 fallback bytes, and 80 accepted-title bytes. A `goals` object opts into the persisted domain, model tools, and same-session driver while forwarding `goals.domain` and `goals.tool` to their owners; omission or `false` leaves the stack absent so headless callers retain one-turn settlement. Set `skills.enabled: false` to omit both the local provider and model-facing skill tool, set `toolBash: false` when another plugin owns the `bash` tool name, and set `toolTasks: false` to retain the task service for foreground producers without exposing `task_output` / `task_list` / `task_kill`. It resolves `dshHome` once through [`@deepseek-ai/dsh-paths`](../../util/paths/README.md) and forwards that absolute value to tool-bash's managed environment and enabled local skill discovery. An absent top-level `dshHome` adopts `skills.local.dshHome`; supplying both with different resolved paths fails loudly. `toolBash.enableRunInBackground` controls only the bundled bash producer; independently loaded producers keep their own config. Workspace instructions register before the skill catalog so their session-prefix message renders first. App packages use `pickSpineConfig()` to copy only these bundle-owned fields.
For example, `{ invariants: { enabled: true, package_allowlist: ['^@deepseek-ai/dsh-'], package_blocklist: ['agent-loop$'] } }` keeps the package-owned companions mounted but suppresses the blocked owner. Blocklist matches override allowlist matches; see [`dsh-invariants`](../../support/invariants/README.md) for regex and lifecycle rules.
@@ -79,5 +79,5 @@ No direct invalidation; the named consumer owns any request-prefix changes.
## Known Limitations and Deferred Work
- **Most of the spine set is fixed in code** — `apply()` always mounts the core services and `tool-bash`; config can omit bundled goals, skills, and task-control tools, but swapping the loop or dropping another spine member means composing a different bundle.
- **Most of the spine set is fixed in code** — `apply()` always mounts the core services; config can omit bundled goals, skills, bash, and task-control tools, but swapping the loop or dropping another spine member means composing a different bundle.
- **The invariant seam and companions remain fixed members** — `invariants.enabled: false` or package filters suppress checks but do not remove the service or companion registrations; Session's always-on validation and freezing are separate.

View File

@@ -31,7 +31,7 @@
@deepseek-ai/dsh-scope/invariant
@deepseek-ai/dsh-agent-loop/invariant
package-owned relational checks
@deepseek-ai/dsh-tool-bash the model-facing bash schema
@deepseek-ai/dsh-tool-bash the model-facing bash schema (unless toolBash=false)
@deepseek-ai/dsh-workspace-context AGENTS.md/CLAUDE.md workspace context loader
@deepseek-ai/dsh-tool-skill session-prefix skill catalog + model-facing loader schema
@deepseek-ai/dsh-tool-tasks task_output/task_list/task_kill schemas + completion notices
@@ -55,11 +55,11 @@
```ts
import type { Config } from '@deepseek-ai/dsh-agent-spine-demo'
// { agents?, maxParallelToolCalls?, persona?, toolOrder?, tools?, dshHome?, sessionTitle?, skills?, workspaceContext, toolBash?, toolTasks?, goals?, invariants? }
// { agents?, maxParallelToolCalls?, includeHarnessIdentity?, persona?, toolOrder?, tools?, dshHome?, sessionTitle?, skills?, workspaceContext, toolBash?, toolTasks?, goals?, invariants? }
// workspaceContext requires { maxBytes } or false; the other owner schemas supply defaults.
```
组合包将每个字段转发给拥有它的子节点:`agents``maxParallelToolCalls` 交给 `agent-loop``agents` 默认为 `[]`,上限在该处默认),因此每个应用提供自己的预创建 agentTUI 和无头应用预创建 `main`ACP 应用则在 `session/new` 按需创建 agent`persona``toolOrder` 交给 `dsh-system-prompt``tools` 交给工具注册表以配置呈现模式;`sessionTitle` 交给后备标题服务;`skills.registry``skills.local``skills.tool` 分别交给 skill 注册表、本地提供方和面向模型的消费方;必填的 `workspaceContext` 选择交给 `dsh-workspace-context``{ maxBytes }` 启用加载,`false` 禁用);`invariants` 交给不变式服务;`toolBash`/`toolTasks` 交给组合包拥有的两个面向模型工具插件。组合包始终挂载 `dsh-llm-retry`,而每个叶节点适配器拥有自己的嵌套 `retryPolicy`。省略 `sessionTitle` 时采用显式示例策略5 个词、40 个后备字节、80 个可接受标题字节。`goals` 对象会选用持久化领域、模型工具和同会话 Goal Round 驱动器,并将 `goals.domain``goals.tool` 转发给各自拥有者;省略或设为 `false` 会让整个栈缺席,使无头调用方继续以单轮次结算。设置 `skills.enabled: false` 会同时省略本地提供方和面向模型的 skill 工具;设置 `toolTasks: false` 会保留供前台生产方使用的任务服务,但不公开 `task_output`/`task_list`/`task_kill`。它对 `dshHome` 只解析一次,解析通过 [`@deepseek-ai/dsh-paths`](../../util/paths/README.md) 完成,并将所得绝对值转发给 tool-bash 的托管环境和已启用的本地 skill 发现。顶层 `dshHome` 缺席时采用 `skills.local.dshHome`;两者同时提供但解析后的路径不同会明确失败。`toolBash.enableRunInBackground` 只控制 bash 生产方;独立加载的生产方保留各自配置。工作区指令先于 skill 目录注册,因此其会话前缀消息先渲染。应用包使用 `pickSpineConfig()`,只复制这些由组合包拥有的字段。
组合包将每个字段转发给拥有它的子节点:`agents``maxParallelToolCalls` 交给 `agent-loop``agents` 默认为 `[]`,上限在该处默认),因此每个应用提供自己的预创建 agentTUI 和无头应用预创建 `main`ACP 应用则在 `session/new` 按需创建 agent`includeHarnessIdentity``persona``toolOrder` 交给 `dsh-system-prompt``tools` 交给工具注册表以配置呈现模式;`sessionTitle` 交给后备标题服务;`skills.registry``skills.local``skills.tool` 分别交给 skill 注册表、本地提供方和面向模型的消费方;必填的 `workspaceContext` 选择交给 `dsh-workspace-context``{ maxBytes }` 启用加载,`false` 禁用);`invariants` 交给不变式服务;`toolBash`/`toolTasks` 交给组合包拥有的两个面向模型工具插件。组合包始终挂载 `dsh-llm-retry`,而每个叶节点适配器拥有自己的嵌套 `retryPolicy`。省略 `sessionTitle` 时采用显式示例策略5 个词、40 个后备字节、80 个可接受标题字节。`goals` 对象会选用持久化领域、模型工具和同会话 Goal Round 驱动器,并将 `goals.domain``goals.tool` 转发给各自拥有者;省略或设为 `false` 会让整个栈缺席,使无头调用方继续以单轮次结算。设置 `skills.enabled: false` 会同时省略本地提供方和面向模型的 skill 工具;当另一个插件拥有 `bash` 工具名时设置 `toolBash: false`设置 `toolTasks: false` 会保留供前台生产方使用的任务服务,但不公开 `task_output`/`task_list`/`task_kill`。它对 `dshHome` 只解析一次,解析通过 [`@deepseek-ai/dsh-paths`](../../util/paths/README.md) 完成,并将所得绝对值转发给 tool-bash 的托管环境和已启用的本地 skill 发现。顶层 `dshHome` 缺席时采用 `skills.local.dshHome`;两者同时提供但解析后的路径不同会明确失败。`toolBash.enableRunInBackground` 只控制内置 bash 生产方;独立加载的生产方保留各自配置。工作区指令先于 skill 目录注册,因此其会话前缀消息先渲染。应用包使用 `pickSpineConfig()`,只复制这些由组合包拥有的字段。
例如,`{ invariants: { enabled: true, package_allowlist: ['^@deepseek-ai/dsh-'], package_blocklist: ['agent-loop$'] } }` 会让包拥有的配套插件保持挂载但抑制被阻止的拥有者。Blocklist 匹配优先于 allowlist 匹配;正则表达式与生命周期规则见 [`dsh-invariants`](../../support/invariants/README.md)。
@@ -79,5 +79,5 @@ YAML include 可以去重配置,却无法拥有 bin 或提供前端入口默
## 已知限制与暂缓事项
- **大部分主干集合固定在代码中**`apply()` 始终挂载核心服务`tool-bash`配置可以省略组合包内的目标、skill 与任务控制工具,但要替换循环或删除其他主干成员,就必须组合另一个组合包。
- **大部分主干集合固定在代码中**`apply()` 始终挂载核心服务配置可以省略组合包内的目标、skill、bash 与任务控制工具,但要替换循环或删除其他主干成员,就必须组合另一个组合包。
- **不变式 seam 与配套插件仍是固定成员**`invariants.enabled: false` 或包筛选器会抑制检查但不会移除服务或配套插件注册Session 始终启用的校验与冻结是另一套机制。

View File

@@ -68,9 +68,9 @@ export interface GoalConfig {
/**
* Bundle config: each field forwarded verbatim to the child that owns it —
* `agents` to the agent loop (an app that pre-creates no agents, like the ACP
* bridge, simply omits it), `persona` and `toolOrder` to the system-prompt
* plugin (the deployment's persona section and the explicit model-facing tool
* order), the `tools` object to the tool registry (its presentation `mode`),
* bridge, simply omits it), `includeHarnessIdentity`, `persona`, and `toolOrder`
* to the system-prompt plugin (the fixed opener, deployment persona, and explicit
* model-facing tool order), the `tools` object to the tool registry (its presentation `mode`),
* `dshHome` to bash environment and local skill discovery, `sessionTitle` to
* the fallback title service, `skills` to the
* skill registry/local provider/tool consumer, `workspaceContext` to the
@@ -83,13 +83,16 @@ export interface GoalConfig {
* workspace context instead requires an explicit byte budget or `false` because
* it changes model-visible input. Producer opt-in stays producer-local:
* `toolBash` configures bash only; independently composed producers keep their
* own config.
* own config. Set `toolBash: false` when another plugin owns the model-facing
* `bash` name.
*/
export interface Config {
/** The agent-loop `agents` list (see dsh-agent-loop's `Config`). */
agents?: AgentLoopConfig['agents']
/** Agent-loop concurrency cap; `1` is serial. */
maxParallelToolCalls?: AgentLoopConfig['maxParallelToolCalls']
/** Whether the system prompt includes the fixed Harness identity (default true). */
includeHarnessIdentity?: SystemPromptConfig['includeHarnessIdentity']
/** The deployment persona (see dsh-system-prompt's `Config`). */
persona?: SystemPromptConfig['persona']
/** The explicit model-facing tool order (see dsh-system-prompt's `Config`). */
@@ -102,10 +105,14 @@ export interface Config {
sessionTitle?: SessionTitleConfig
/** Workspace-context loader controls with an explicit byte budget; set `false` for hermetic prompts. */
workspaceContext: workspaceContext.Config | false
/** Skill registry, local provider, and model-facing consumer config. */
/**
* Skill registry, local provider, and model-facing consumer config.
* Skills use `enabled` because one nested config controls a provider stack;
* single model-tool plugins use `Config | false` to disable that one consumer.
*/
skills?: SkillConfig
/** Model-facing bash tool config, including this producer's background opt-in. */
toolBash?: toolBash.Config
/** Model-facing bash tool config, or false when another plugin owns `bash`. */
toolBash?: toolBash.Config | false
/** Generic background-task controls; set false to keep the task service without model-facing task tools. */
toolTasks?: toolTasks.Config | false
/** Global enablement and package-name filters for invariant companions. */
@@ -127,7 +134,8 @@ export const SessionTitleConfigSchema: z<SessionTitleConfig> = SessionTitleServi
.default(EXAMPLE_SESSION_TITLE_CONFIG)
/** The bash-tool config schema exported for app packages that forward `toolBash`. */
export const ToolBashConfigSchema: z<toolBash.Config> = toolBash.Config
export const ToolBashConfigSchema: z<toolBash.Config | false> =
z.union([z.const(false), toolBash.Config])
/** The task-control-tool config schema exported for app packages that forward `toolTasks`. */
export const ToolTasksConfigSchema: z<toolTasks.Config> = toolTasks.Config
@@ -163,6 +171,7 @@ export const Config = z.intersect([
export function pickSpineConfig(config: Omit<Config, 'agents'>): Omit<Config, 'agents'> {
return {
...config.maxParallelToolCalls !== undefined ? { maxParallelToolCalls: config.maxParallelToolCalls } : {},
...config.includeHarnessIdentity !== undefined ? { includeHarnessIdentity: config.includeHarnessIdentity } : {},
...config.persona !== undefined ? { persona: config.persona } : {},
...config.toolOrder !== undefined ? { toolOrder: config.toolOrder } : {},
...config.tools !== undefined ? { tools: config.tools } : {},
@@ -201,6 +210,7 @@ export function apply(ctx: Context, config: Config): void {
ctx.plugin(SessionTitleService, config.sessionTitle ?? EXAMPLE_SESSION_TITLE_CONFIG)
// Owner schemas resolve defaults; forward toolOrder only when explicitly set.
ctx.plugin(SystemPrompt, {
includeHarnessIdentity: config.includeHarnessIdentity ?? true,
persona: config.persona ?? '',
...config.toolOrder !== undefined ? { toolOrder: config.toolOrder } : {},
})
@@ -223,7 +233,9 @@ export function apply(ctx: Context, config: Config): void {
ctx.plugin(agentInvariant)
ctx.plugin(scopeInvariant)
ctx.plugin(agentLoopInvariant)
ctx.plugin(toolBash, Object.assign({}, config.toolBash, { dshHome }))
if (config.toolBash !== false) {
ctx.plugin(toolBash, Object.assign({}, config.toolBash, { dshHome }))
}
if (config.workspaceContext !== false) {
ctx.plugin(workspaceContext, config.workspaceContext)
}

View File

@@ -4,7 +4,7 @@ import { join } from 'node:path'
import { tmpdir } from 'node:os'
import { Context } from 'cordis'
import Loader from '@cordisjs/plugin-loader'
import { TOOL_ORDER_REST } from '@deepseek-ai/dsh-system-prompt'
import { renderPrompt, TOOL_ORDER_REST } from '@deepseek-ai/dsh-system-prompt'
import * as agentCore from '../src/index.ts'
import { agentEvents, type Agent } from '@deepseek-ai/dsh-agent'
import { SessionId } from '@deepseek-ai/dsh-session'
@@ -654,9 +654,27 @@ describe('dsh-agent-spine-demo bundle', () => {
await ctx.fiber.dispose()
})
it('can omit the bundled bash tool and Harness identity for a compatibility deployment', async () => {
const ctx = await mount({
includeHarnessIdentity: false,
persona: 'You are a helpful software engineer assistant.',
workspaceContext: false,
skills: { enabled: false },
toolBash: false,
toolTasks: false,
}, true)
expect(ctx.tools.schemas()).toEqual([])
expect(renderPrompt(await ctx.systemPrompt.assemble()))
.toBe('You are a helpful software engineer assistant.')
await ctx.fiber.dispose()
})
it('picks shared spine config without leaking front-door fields', () => {
const appConfig = {
model: 'front-door-only',
includeHarnessIdentity: false,
persona: 'You are merged.',
toolOrder: ['zulu'],
tools: { mode: 'native' as const },
@@ -670,6 +688,7 @@ describe('dsh-agent-spine-demo bundle', () => {
}
expect(agentCore.pickSpineConfig(appConfig)).toEqual({
includeHarnessIdentity: appConfig.includeHarnessIdentity,
persona: appConfig.persona,
toolOrder: appConfig.toolOrder,
tools: appConfig.tools,

View File

@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write packages/fs/README.md
README.md: 4d954455ea920be4882530bcfe90b48a364c29b5
README.zh.md: ed154cbbaf7c7f45c42c17799957254fee918153
README.md: b5e0ac9d1c0c550eb372b8a66fc6358711fddc07
README.zh.md: ee64a617aa0d9549bcaf00b20821a6d59c6673c8

View File

@@ -12,6 +12,7 @@ The filesystem stack: a provider seam (text IO + atomic mutation with an optiona
| `fs-policy/` | Policy gate plugin: observed-state + read-before-edit + version-guarded write/edit, via the `fs/*` event gate | (no service — `fs/*` listeners) |
| `tool-fs/` | Model-facing `read`/`write`/`edit` tools AND the executor (reads via `ctx.fs`, owns read windowing, dispatches `fs/*`); preserves filesystem semantics for session-cwd-relative paths and advertises sandbox escalation fields when the mounted `ctx.fs` confines | (registers on `ctx.tools`) |
| `tool-fs-search/` | Model-facing `glob`/`grep` discovery tools when `rg` is available on the bash executor `PATH`, backed by fixed ripgrep commands through `ctx.bash`, NOT by `ctx.fs` provider methods | (registers on `ctx.tools`) |
| `tool-str-replace-editor/` | Model-facing `str_replace_editor` with view/create/unique literal replace/line insert operations over `ctx.fs` | (registers on `ctx.tools`) |
The interface lives at `fs/fs/`. A sandboxed, remote, or project-scoped filesystem backend can replace `fs-local` without touching the seam, the policy gate, or the model-facing tool schemas — `fs-sandbox` is the first such replacement (an in-process path fence over the shared sandbox mode; see [the cross-family fs sandbox Agent Note](../../.agents/notes/implemented/feature/2026-07-14-cross-family-fs-sandbox.md)). The policy (`fs-policy/`) is a plugin that participates only through the `fs/*` event gate, not a service the tool injects — so dropping it gracefully loses the policy and leaves the unconstrained bare provider rather than breaking the tool. A deployment that loads `tool-fs/` is expected to also load it. The mode fence and the read-before-edit gate are orthogonal and compose. Discovery (`tool-fs-search/`) deliberately does NOT extend the provider seam: search is a process-backed `rg` workflow on the bash executor, so filesystem backends stay free of a universal search contract; its tools register only when that executor can find `rg`, and its results are follow-up-readable when the bash workdir and the `read` root are the same workspace (the co-located deployment its README documents).

View File

@@ -12,6 +12,7 @@
| `fs-policy/` | 策略门禁插件:通过 `fs/*` 事件门禁提供已观察状态、编辑前读取和版本防护的写入/编辑 | (无服务,仅有 `fs/*` 监听器) |
| `tool-fs/` | 面向模型的 `read`/`write`/`edit` 工具以及执行器(通过 `ctx.fs` 读取,拥有读取窗口逻辑,分派 `fs/*`);为会话 cwd 相对路径保留文件系统语义,并在已挂载的 `ctx.fs` 实施约束时声明沙箱升权字段 | (注册到 `ctx.tools` |
| `tool-fs-search/` | 面向模型的 `glob`/`grep` 发现工具;当 `rg` 位于 bash 执行器 `PATH` 上时注册,通过 `ctx.bash` 运行固定 ripgrep 命令,而不是使用 `ctx.fs` 提供方方法 | (注册到 `ctx.tools` |
| `tool-str-replace-editor/` | 基于 `ctx.fs` 提供查看/创建/唯一字面量替换/按行插入的模型可见 `str_replace_editor` | (注册到 `ctx.tools` |
接口位于 `fs/fs/`。沙箱化、远程或限定项目作用域的文件系统后端可以替换 `fs-local`,而无需更改 seam、策略门禁或面向模型的工具 schema`fs-sandbox` 是第一个这样的替代实现(基于共享沙箱模式的进程内路径围栏;见[跨能力族 fs 沙箱 Agent Noteagent 决策记录)](../../.agents/notes/implemented/feature/2026-07-14-cross-family-fs-sandbox.md))。策略(`fs-policy/`)是一个只通过 `fs/*` 事件门禁参与的插件,不是工具注入的服务;因此移除它只会使策略失效,留下不受约束的裸提供方,而不会破坏工具。加载 `tool-fs/` 的部署也应加载该插件。模式围栏与编辑前读取门禁彼此正交,可以组合。发现(`tool-fs-search/`)有意不扩展提供方 seam搜索是在 bash 执行器上运行 `rg`、基于进程的工作流,因此文件系统后端无需承担通用搜索契约;只有当执行器能找到 `rg` 时,其工具才会注册。如果 bash 工作目录与 `read` 根目录是同一工作区,其结果便可供后续读取,这也是其 README 所述的共置部署。

View File

@@ -0,0 +1,6 @@
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write packages/fs/tool-str-replace-editor/README.md
README.md: 97e9e0ab9ade7c7241c1aac3e2489e055d01ff8f
README.zh.md: 48358eb3c9d81ddad6a83c4ff3ef0cf6542096b1

View File

@@ -0,0 +1,52 @@
# @deepseek-ai/dsh-tool-str-replace-editor
English | [中文](README.zh.md)
Standalone model-facing `str_replace_editor` over `ctx.fs`. It can be composed with persistent Bash, one-shot Bash, sandboxed Bash, or another terminal surface.
## Config
| Key | Default | Meaning |
|---|---:|---|
| `maxOutputChars` | `16000` | Prefix characters retained for file and directory views. |
| `description` | Editor command guide | Model-facing tool description. |
## Tool
The schema provides `view`, `create`, `str_replace`, and `insert` over absolute paths. File views use one-based line numbers and preserve content tabs, so displayed text remains valid literal replacement input; directory views omit hidden, dependency, and Python-cache entries and descend two levels. Replacement requires one unique literal match and reports errors only in the public `old_str` vocabulary. Insert follows the selected zero-based insertion boundary without adding an implicit trailing newline. Mutations preserve tabs outside the requested edit.
## Model Experience
### Tool schema
#### What the model sees
The generated [`str_replace_editor` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-str-replace-editor), including the configured `description`. The plugin contributes no standalone system-prompt section.
#### Token effect
Fixed schema cost while `str_replace_editor` is visible.
#### KV Cache effect
Prefix-stable while the configured description and schema remain unchanged.
### Tool results
#### What the model sees
Views return numbered text or a shallow directory listing. Calls expose file locations, and create/replace calls expose diff cards to presentation surfaces. Mutations return concise confirmations. Long views keep their prefix and append a clipping notice.
#### Token effect
Data-dependent and bounded by `maxOutputChars` plus the fixed clipping notice.
#### KV Cache effect
Append-only tool results follow the reusable request prefix.
## Known Limitations and Deferred Work
- Operations target UTF-8 text; binary files are unsupported.
- `str_replace` intentionally rejects zero or multiple matches and has no `replace_all` argument.
- Every mutation goes through `fs/write-intent` or `fs/edit-intent`, resolves the current session sandbox policy, and delegates enforcement to the mounted filesystem and policy plugins.

View File

@@ -0,0 +1,52 @@
# @deepseek-ai/dsh-tool-str-replace-editor
[English](README.md) | 中文
基于 `ctx.fs` 的独立模型可见 `str_replace_editor`。它可与持久 Bash、一次性 Bash、沙箱 Bash 或其他终端表面组合。
## 配置
| 键 | 默认值 | 含义 |
|---|---:|---|
| `maxOutputChars` | `16000` | 文件和目录查看结果保留的前缀字符数。 |
| `description` | 编辑器命令指南 | 面向模型的工具描述。 |
## 工具
Schema 提供针对绝对路径的 `view``create``str_replace``insert`。文件查看使用从一开始的行号,并保留内容中的制表符,因此显示的文本仍可作为有效的字面量替换输入;目录查看忽略隐藏、依赖与 Python 缓存条目并下探两层。替换要求字面量唯一匹配,错误只使用公开的 `old_str` 词汇。插入遵循所选的零基插入边界,不会隐式补尾换行。修改操作会保留请求编辑范围之外的制表符。
## 模型体验
### 工具 schema
#### 模型所见
生成的 [`str_replace_editor` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-str-replace-editor),其中包含配置的 `description`。本插件不贡献独立系统提示词段。
#### Token 影响
`str_replace_editor` 可见时产生固定的 schema 成本。
#### KV Cache 影响
配置的描述与 schema 不变时前缀稳定。
### 工具结果
#### 模型所见
查看操作返回带行号文本或浅层目录列表。调用会向展示层提供文件位置,创建/替换还会提供 diff 卡片。修改操作返回简洁确认。长查看结果保留前缀并追加截断提示。
#### Token 影响
随数据变化,并受 `maxOutputChars` 与固定截断提示约束。
#### KV Cache 影响
工具结果以追加方式位于可复用请求前缀之后。
## 已知限制与延后工作
- 操作面向 UTF-8 文本,不支持二进制文件。
- `str_replace` 刻意拒绝零匹配或多匹配,且没有 `replace_all` 参数。
- 每个修改操作都会经过 `fs/write-intent``fs/edit-intent`,解析当前 session 的沙箱策略,并交由挂载的文件系统与策略插件执行。

View File

@@ -0,0 +1,54 @@
{
"name": "@deepseek-ai/dsh-tool-str-replace-editor",
"description": "Model-facing view, create, literal replace, and line insert tool over the Harness filesystem service",
"version": "0.0.1",
"private": true,
"type": "module",
"main": "lib/index.js",
"types": "lib/types/index.d.ts",
"exports": {
".": {
"types": "./lib/types/index.d.ts",
"default": "./lib/index.js"
},
"./invariant": {
"types": "./lib/types/invariant.d.ts",
"default": "./lib/invariant.js"
},
"./package.json": "./package.json"
},
"files": [
"lib/index.js",
"lib/invariant.js",
"lib/types/**/*.d.ts",
"lib/types/**/*.d.ts.map",
"src"
],
"license": "BSD-3-Clause",
"peerDependencies": {
"@deepseek-ai/dsh-fs": "^0.0.1",
"@deepseek-ai/dsh-invariants": "^0.0.1",
"@deepseek-ai/dsh-sandbox": "^0.0.1",
"@deepseek-ai/dsh-sandbox-policy": "^0.0.1",
"@deepseek-ai/dsh-tools": "^0.0.1",
"cordis": "^4.0.0-rc.7"
},
"dependencies": {
"schemastery": "^3.18.0"
},
"devDependencies": {
"@deepseek-ai/dsh-agent": "workspace:^",
"@deepseek-ai/dsh-fs": "workspace:^",
"@deepseek-ai/dsh-fs-local": "workspace:^",
"@deepseek-ai/dsh-fs-policy": "workspace:^",
"@deepseek-ai/dsh-fs-sandbox": "workspace:^",
"@deepseek-ai/dsh-invariants": "workspace:^",
"@deepseek-ai/dsh-llm": "workspace:^",
"@deepseek-ai/dsh-sandbox": "workspace:^",
"@deepseek-ai/dsh-sandbox-policy": "workspace:^",
"@deepseek-ai/dsh-session": "workspace:^",
"@deepseek-ai/dsh-system-prompt": "workspace:^",
"@deepseek-ai/dsh-tools": "workspace:^",
"cordis": "^4.0.0-rc.7"
}
}

View File

@@ -0,0 +1,522 @@
/**
* Model-facing `str_replace_editor` over the Harness filesystem seam.
* @module @deepseek-ai/dsh-tool-str-replace-editor
*/
import { isAbsolute } from 'node:path'
import type { Context } from 'cordis'
import z from 'schemastery'
import { FsError } from '@deepseek-ai/dsh-fs'
import type { FsInfo, FsTarget, FsWriteIntent } from '@deepseek-ai/dsh-fs'
import { sandboxDenialMarker } from '@deepseek-ai/dsh-sandbox'
import type { SandboxExecutionPolicy } from '@deepseek-ai/dsh-sandbox'
import type { SandboxPolicyService } from '@deepseek-ai/dsh-sandbox-policy'
import { defineTool } from '@deepseek-ai/dsh-tools'
import type { ToolCallView, ToolRunContext } from '@deepseek-ai/dsh-tools'
const TRUNCATED_MESSAGE = '<response clipped><NOTE>To save on context only part of this file has been shown to you. You should retry this tool after you have searched inside the file with `grep -n` in order to find the line numbers of what you are looking for.</NOTE>'
const DEFAULT_DESCRIPTION = `
Custom editing tool for viewing, creating and editing files
* State is persistent across command calls and discussions with the user
* If \`path\` is a file, \`view\` displays the result of applying \`cat -n\`. If \`path\` is a directory, \`view\` lists non-hidden files and directories up to 2 levels deep
* The \`create\` command cannot be used if the specified \`path\` already exists as a file
* If a \`command\` generates a long output, it will be truncated and marked with \`<response clipped>\`
Notes for using the \`str_replace\` command:
* The \`old_str\` parameter should match EXACTLY one or more consecutive lines from the original file. Be mindful of whitespaces!
* If the \`old_str\` parameter is not unique in the file, the replacement will not be performed. Make sure to include enough context in \`old_str\` to make it unique
* The \`new_str\` parameter should contain the edited lines that should replace the \`old_str\`
`.trim()
function maybeTruncate(content: string, maxOutputChars: number): string {
return content.length <= maxOutputChars
? content
: content.slice(0, maxOutputChars) + TRUNCATED_MESSAGE
}
function codepointCompare(left: string, right: string): number {
return left < right ? -1 : left > right ? 1 : 0
}
function matchOffsets(content: string, search: string): number[] {
const offsets: number[] = []
let offset = 0
while (true) {
const match = content.indexOf(search, offset)
if (match < 0) return offsets
offsets.push(match)
offset = match + search.length
}
}
function lineNumbersAt(content: string, offsets: readonly number[]): number[] {
let line = 1
let cursor = 0
return offsets.map((offset) => {
while (cursor < offset) {
if (content[cursor] === '\n') line += 1
cursor += 1
}
return line
})
}
class MutationPolicy {
private readonly policy: SandboxPolicyService | undefined
constructor(ctx: Context) {
this.policy = ctx.fs.sandboxMode === undefined ? undefined : ctx.get('sandboxPolicy')
if (ctx.fs.sandboxMode !== undefined && this.policy === undefined) {
throw new Error('tool-str-replace-editor: the mounted filesystem confines but ctx.sandboxPolicy is missing')
}
}
resolve(exec: ToolRunContext): SandboxExecutionPolicy | undefined {
return this.policy?.resolve({
...exec.agent === undefined ? {} : { session: exec.agent.session },
})
}
mapError(error: unknown, policy: SandboxExecutionPolicy | undefined): unknown {
if (!(error instanceof FsError) || error.code !== 'FS_SANDBOX_DENIED') return error
const mode = (policy as SandboxExecutionPolicy).mode
return new FsError(sandboxDenialMarker(mode), 'FS_SANDBOX_DENIED', { cause: error })
}
}
async function resolveTarget(
ctx: Context,
path: string,
signal: AbortSignal,
): Promise<FsTarget> {
if (path.trim().length === 0) throw new Error('path must be a non-empty string')
if (!isAbsolute(path)) {
throw new Error(`The path ${path} is not an absolute path, it should start with \`/\`. Maybe you meant /${path}?`)
}
return ctx.fs.resolve(path, { signal })
}
async function statExisting(
ctx: Context,
target: FsTarget,
command: 'view' | 'str_replace' | 'insert',
exec: ToolRunContext,
): Promise<FsInfo> {
const info = await ctx.fs.stat(target, exec.signal)
if (info === undefined) {
throw new FsError(
`The path ${target.displayPath} does not exist. Please provide a valid path.`,
'FS_NOT_FOUND',
)
}
if (info.type === 'directory' && command !== 'view') {
throw new FsError(
`The path ${target.displayPath} is a directory and only the \`view\` command can be used on directories`,
'FS_NOT_REGULAR_FILE',
)
}
return info
}
function requiredForCommand(
value: string | undefined,
parameter: string,
command: string,
allowEmpty = true,
): string {
if (value === undefined) throw new Error(`Parameter \`${parameter}\` is required for command: ${command}`)
if (!allowEmpty && value.length === 0) {
throw new Error(`Parameter \`${parameter}\` is empty for command: ${command}`)
}
return value
}
function formatFileView(
path: string,
content: string,
maxOutputChars: number,
viewRange?: number[],
): string {
const allLines = content.split('\n')
let lines = allLines
let initialLine = 1
let finalLine: number | undefined
let prompt = `Here's the content of ${path} with line numbers (which has a total of ${allLines.length} lines)`
if (viewRange !== undefined) {
const [requestedInitialLine, requestedFinalLine] = viewRange
if (
viewRange.length !== 2
|| requestedInitialLine === undefined
|| requestedFinalLine === undefined
|| !viewRange.every(Number.isInteger)
) {
throw new Error('Invalid `view_range`. It should be a list of two integers.')
}
initialLine = requestedInitialLine
finalLine = requestedFinalLine
if (initialLine < 1 || initialLine > allLines.length) {
throw new Error(
`Invalid \`view_range\`: [${viewRange.join(', ')}]. Its first element \`${initialLine}\` should be within the range of lines of the file: [1, ${allLines.length}]`,
)
}
if (finalLine > allLines.length) {
throw new Error(
`Invalid \`view_range\`: [${viewRange.join(', ')}]. Its second element \`${finalLine}\` should be smaller than the number of lines in the file: \`${allLines.length}\``,
)
}
if (finalLine !== -1 && finalLine < initialLine) {
throw new Error(
`Invalid \`view_range\`: [${viewRange.join(', ')}]. Its second element \`${finalLine}\` should be larger or equal than its first \`${initialLine}\``,
)
}
lines = finalLine === -1
? allLines.slice(initialLine - 1)
: allLines.slice(initialLine - 1, finalLine)
prompt += ` with view_range=[${initialLine}, ${finalLine}]`
}
const numbered = lines
.map((line, index) => `${String(initialLine + index).padStart(6, ' ')} ${line}`)
.join('\n')
return maybeTruncate(`${prompt}:\n${numbered}\n`, maxOutputChars)
}
async function listDirectory(
ctx: Context,
target: FsTarget,
maxOutputChars: number,
exec: ToolRunContext,
): Promise<string> {
async function visit(dir: FsTarget, depth: number): Promise<string[]> {
const entries = await ctx.fs.listDir(dir, exec.signal)
const rows: string[] = []
for (const entry of entries.filter(candidate =>
!candidate.name.startsWith('.')
&& candidate.name !== 'node_modules'
&& candidate.name !== '__pycache__')) {
const type = entry.type === 'directory' ? 'd' : entry.type === 'file' ? 'f' : '?'
rows.push(`${type}\t${entry.target.displayPath}`)
if (entry.type === 'directory' && depth < 2) {
rows.push(...await visit(entry.target, depth + 1))
}
}
return rows
}
const rows = [`d\t${target.displayPath}`, ...await visit(target, 1)]
rows.sort((left, right) => {
const leftPath = left.slice(left.indexOf('\t') + 1)
const rightPath = right.slice(right.indexOf('\t') + 1)
return codepointCompare(leftPath, rightPath)
})
const listing = maybeTruncate(rows.join('\n') + '\n', maxOutputChars)
return `Here're the files and directories up to 2 levels deep in ${target.displayPath}, excluding hidden items, node_modules, and Python cache directories:\n${listing}\n`
}
async function viewPath(
ctx: Context,
path: string,
viewRange: number[] | undefined,
maxOutputChars: number,
exec: ToolRunContext,
): Promise<string> {
const target = await resolveTarget(ctx, path, exec.signal)
const info = await statExisting(ctx, target, 'view', exec)
if (info.type === 'directory') {
if (viewRange !== undefined) {
throw new Error('The `view_range` parameter is not allowed when `path` points to a directory.')
}
return listDirectory(ctx, target, maxOutputChars, exec)
}
if (info.type !== 'file') {
throw new FsError(`cannot view "${target.displayPath}": not a regular file or directory`, 'FS_NOT_REGULAR_FILE')
}
const content = await ctx.fs.readText(target, exec.signal)
ctx.emit('fs/observed', target, info.version, exec)
return formatFileView(target.displayPath, content, maxOutputChars, viewRange)
}
async function createFile(
ctx: Context,
policy: MutationPolicy,
path: string,
fileText: string | undefined,
exec: ToolRunContext,
): Promise<string> {
const content = requiredForCommand(fileText, 'file_text', 'create')
const sandboxPolicy = policy.resolve(exec)
const target = await resolveTarget(ctx, path, exec.signal)
if (await ctx.fs.stat(target, exec.signal) !== undefined) {
throw new Error(`File already exists at: ${target.displayPath}. Cannot overwrite files using command \`create\`.`)
}
const intent = await ctx.waterfall(
'fs/write-intent',
target,
exec,
() => ({ kind: 'createIfAbsent' } as const),
)
let outcome
try {
outcome = await ctx.fs.writeText(
target,
content,
intent,
exec.signal,
sandboxPolicy,
)
} catch (error: unknown) {
throw policy.mapError(error, sandboxPolicy)
}
ctx.emit('fs/observed', target, outcome.version, exec)
return `New file created successfully at: ${target.displayPath}`
}
async function replaceInFile(
ctx: Context,
policy: MutationPolicy,
path: string,
oldStr: string | undefined,
newStr: string | undefined,
exec: ToolRunContext,
): Promise<string> {
const sandboxPolicy = policy.resolve(exec)
const target = await resolveTarget(ctx, path, exec.signal)
const intent = await ctx.waterfall('fs/edit-intent', target, exec, () => undefined)
const oldValue = requiredForCommand(oldStr, 'old_str', 'str_replace', false)
const newValue = newStr ?? ''
const info = await statExisting(ctx, target, 'str_replace', exec)
if (info.type !== 'file') {
throw new FsError(`cannot edit "${target.displayPath}": not a regular file`, 'FS_NOT_REGULAR_FILE')
}
const before = await ctx.fs.readText(target, exec.signal)
const offsets = matchOffsets(before, oldValue)
const offset = offsets[0]
if (offset === undefined) {
throw new FsError(
`No replacement was performed, old_str \`${oldValue}\` did not appear verbatim in ${target.displayPath}.`,
'FS_EDIT_NOT_FOUND',
)
}
if (offsets.length > 1) {
const lines = lineNumbersAt(before, offsets)
throw new FsError(
`No replacement was performed. Multiple occurrences of old_str \`${oldValue}\` in lines [${lines.join(', ')}]. Please ensure it is unique`,
'FS_AMBIGUOUS_EDIT',
)
}
let outcome
try {
outcome = await ctx.fs.writeText(
target,
before.slice(0, offset) + newValue + before.slice(offset + oldValue.length),
intent === undefined
? { kind: 'replaceIfVersion', version: info.version }
: { kind: 'replaceIfVersion', version: intent.version },
exec.signal,
sandboxPolicy,
)
} catch (error: unknown) {
throw policy.mapError(error, sandboxPolicy)
}
ctx.emit('fs/observed', target, outcome.version, exec)
return `The file ${target.displayPath} has been edited successfully.`
}
async function insertInFile(
ctx: Context,
policy: MutationPolicy,
path: string,
insertLine: number | undefined,
newStr: string | undefined,
exec: ToolRunContext,
): Promise<string> {
if (insertLine === undefined) throw new Error('Parameter `insert_line` is required for command: insert')
const value = requiredForCommand(newStr, 'new_str', 'insert')
const sandboxPolicy = policy.resolve(exec)
const target = await resolveTarget(ctx, path, exec.signal)
const intent = await ctx.waterfall('fs/edit-intent', target, exec, () => undefined)
const info = await statExisting(ctx, target, 'insert', exec)
if (info.type !== 'file') {
throw new FsError(`cannot insert into "${target.displayPath}": not a regular file`, 'FS_NOT_REGULAR_FILE')
}
const before = await ctx.fs.readText(target, exec.signal)
const lines = before.split('\n')
if (!Number.isInteger(insertLine) || insertLine < 0 || insertLine > lines.length) {
throw new Error(
`Invalid \`insert_line\` parameter: ${insertLine}. It should be within the range of lines of the file: [0, ${lines.length}]`,
)
}
const after = [
...lines.slice(0, insertLine),
...value.split('\n'),
...lines.slice(insertLine),
].join('\n')
const expected: FsWriteIntent = intent === undefined
? { kind: 'replaceIfVersion', version: info.version }
: { kind: 'replaceIfVersion', version: intent.version }
let outcome
try {
outcome = await ctx.fs.writeText(target, after, expected, exec.signal, sandboxPolicy)
} catch (error: unknown) {
throw policy.mapError(error, sandboxPolicy)
}
ctx.emit('fs/observed', target, outcome.version, exec)
return `The file ${target.displayPath} has been edited successfully.`
}
interface ResolvedConfig {
maxOutputChars: number
description: string
}
function presentEditorCall(args: {
command: 'view' | 'create' | 'str_replace' | 'insert'
path: string
file_text?: string
insert_line?: number
new_str?: string
old_str?: string
}): ToolCallView {
switch (args.command) {
case 'view':
return {
card: 'generic',
title: `view ${args.path}`,
kind: 'read',
locations: [{ path: args.path }],
}
case 'create':
return {
card: 'diff',
title: `create ${args.path}`,
diffs: [{ path: args.path, oldText: null, newText: args.file_text ?? '' }],
locations: [{ path: args.path }],
}
case 'str_replace':
return {
card: 'diff',
title: `str_replace ${args.path}`,
diffs: [{
path: args.path,
oldText: args.old_str ?? null,
newText: args.new_str ?? '',
}],
locations: [{ path: args.path }],
}
case 'insert':
return {
card: 'generic',
title: `insert ${args.path}`,
kind: 'edit',
locations: [{
path: args.path,
...args.insert_line === undefined ? {} : { line: Math.max(1, args.insert_line + 1) },
}],
}
}
}
/** Register the model-facing `str_replace_editor` tool. */
function registerStrReplaceEditor(ctx: Context, config: ResolvedConfig): void {
const policy = new MutationPolicy(ctx)
ctx.tools.register(defineTool({
name: 'str_replace_editor',
description: config.description,
parameters: {
command: {
type: 'string',
required: true,
enum: ['view', 'create', 'str_replace', 'insert'],
description: 'The commands to run. Allowed options are: `view`, `create`, `str_replace`, `insert`.',
},
path: {
type: 'string',
required: true,
description: 'Absolute path to file or directory, e.g. `/repo/file.py` or `/repo`.',
},
file_text: {
type: 'string',
description: 'Required parameter of `create` command, with the content of the file to be created.',
},
insert_line: {
type: 'integer',
description: 'Required parameter of `insert` command. The `new_str` will be inserted AFTER the line `insert_line` of `path`.',
},
new_str: {
type: 'string',
description: 'Optional parameter of `str_replace` command containing the new string (if not given, no string will be added). Required parameter of `insert` command containing the string to insert.',
},
old_str: {
type: 'string',
description: 'Required parameter of `str_replace` command containing the string in `path` to replace.',
},
view_range: {
type: 'array',
items: { type: 'integer' },
description: 'Optional parameter of `view` command when `path` points to a file. If none is given, the full file is shown. If provided, the file will be shown in the indicated line number range, e.g. [11, 12] will show lines 11 and 12. Indexing at 1 to start. Setting `[start_line, -1]` shows all lines from `start_line` to the end of the file.',
},
},
output: {
schema: { type: 'string' },
render: (_args, value) => [{ type: 'text', text: value }],
},
async execute(args, exec) {
switch (args.command) {
case 'view':
return viewPath(ctx, args.path, args.view_range, config.maxOutputChars, exec)
case 'create':
return createFile(ctx, policy, args.path, args.file_text, exec)
case 'str_replace':
return replaceInFile(
ctx,
policy,
args.path,
args.old_str,
args.new_str,
exec,
)
case 'insert':
return insertInFile(
ctx,
policy,
args.path,
args.insert_line,
args.new_str,
exec,
)
}
},
presentCall: presentEditorCall,
}))
}
export const name = 'tool-str-replace-editor'
export const inject = ['tools', 'fs']
/** Configuration for the string-replacement editor tool. */
export interface Config {
/** Maximum returned view characters before clipping (default 16000). */
maxOutputChars?: number
/** Model-facing tool description. */
description?: string
}
/** Runtime configuration schema for the string-replacement editor tool. */
export const Config: z<Config> = z.object({
maxOutputChars: z.number().default(16_000),
description: z.string().default(DEFAULT_DESCRIPTION),
})
/** Register one `str_replace_editor` tool over `ctx.fs`. */
export function apply(ctx: Context, config: Config): void {
const resolved: ResolvedConfig = {
maxOutputChars: config.maxOutputChars ?? 16_000,
description: config.description ?? DEFAULT_DESCRIPTION,
}
if (!Number.isSafeInteger(resolved.maxOutputChars) || resolved.maxOutputChars <= 0) {
throw new Error('tool-str-replace-editor: maxOutputChars must be a positive safe integer')
}
if (resolved.description.trim().length === 0) {
throw new Error('tool-str-replace-editor: description must be non-empty')
}
registerStrReplaceEditor(ctx, resolved)
}

View File

@@ -0,0 +1,30 @@
/**
* Package-owned invariant companion for `@deepseek-ai/dsh-tool-str-replace-editor`.
* @module @deepseek-ai/dsh-tool-str-replace-editor/invariant
*/
/* jscpd:ignore-start */
import type { Context } from 'cordis'
import type { InvariantInstaller } from '@deepseek-ai/dsh-invariants'
const PACKAGE_NAME = '@deepseek-ai/dsh-tool-str-replace-editor'
/** Cordis companion plugin name. */
export const name = 'tool-str-replace-editor-invariant'
/** Service required before the companion can reserve package ownership. */
export const inject = ['invariants']
/**
* No runtime invariant: the tool adapter owns no independent durable state;
* filesystem mutation relations stay with the provider and policy plugins.
*/
const install: InvariantInstaller = () => {}
/**
* Register this package's invariant companion.
* @param ctx - Cordis context carrying the invariant service.
* @returns the installed registration's disposer after setup succeeds.
*/
export const apply = (ctx: Context): Promise<() => void> =>
Promise.resolve(ctx.invariants.register(PACKAGE_NAME, install))
/* jscpd:ignore-end */

View File

@@ -0,0 +1,547 @@
import { mkdtemp, mkdir, readFile, rm, writeFile } from 'node:fs/promises'
import { tmpdir } from 'node:os'
import { join } from 'node:path'
import { afterEach, describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import { FsVersion } from '@deepseek-ai/dsh-fs'
import { CallId } from '@deepseek-ai/dsh-llm'
import { Session, SessionId } from '@deepseek-ai/dsh-session'
import AgentRegistry from '@deepseek-ai/dsh-agent'
import type { Agent } from '@deepseek-ai/dsh-agent'
import LocalFileSystem from '@deepseek-ai/dsh-fs-local'
import * as FsPolicy from '@deepseek-ai/dsh-fs-policy'
import SandboxedFileSystem from '@deepseek-ai/dsh-fs-sandbox'
import SandboxPolicy from '@deepseek-ai/dsh-sandbox-policy'
import SystemPrompt from '@deepseek-ai/dsh-system-prompt'
import ToolRegistry from '@deepseek-ai/dsh-tools'
import * as ToolStrReplaceEditor from '@deepseek-ai/dsh-tool-str-replace-editor'
const contexts: Context[] = []
const roots: string[] = []
let callNumber = 0
afterEach(async () => {
for (const ctx of contexts.splice(0)) await ctx.fiber.dispose()
for (const root of roots.splice(0)) await rm(root, { recursive: true, force: true })
})
function agent(ctx: Context, cwd: string): Agent {
const id = SessionId(`str-replace-editor-owner-${callNumber}`)
const scope = ctx.plugin(() => {})
const value: Agent = {
id,
options: {},
session: new Session(id, [], { version: 0, id, createdAt: 0, cwd }),
status: 'idle',
acceptsNextStep: false,
ctx: scope.ctx,
followup: () => {},
steer: () => {},
inject: () => {},
send: () => {},
updateInbox: () => 'not-found',
cancel() {},
whenIdle: () => Promise.resolve(),
}
ctx.agents.register(value)
return value
}
function text(result: { content: { type: string; text?: string }[] }): string {
return result.content.filter(block => block.type === 'text').map(block => block.text).join('')
}
function call(ctx: Context, owner: Agent | undefined, args: unknown) {
return ctx.tools.execute({
signal: new AbortController().signal,
callId: CallId(`str-replace-editor-${++callNumber}`),
name: 'str_replace_editor',
arguments: args,
...owner === undefined ? {} : { agent: owner },
})
}
async function setup(
config: ToolStrReplaceEditor.Config = {},
options: { fsPolicy?: boolean; sandboxMode?: 'read-only' | 'workspace-write' | 'danger-full-access' } = {},
) {
const root = await mkdtemp(join(tmpdir(), 'dsh-tool-str-replace-editor-'))
roots.push(root)
const ctx = new Context()
contexts.push(ctx)
await ctx.plugin(SystemPrompt)
await ctx.plugin(ToolRegistry)
await ctx.plugin(AgentRegistry)
if (options.sandboxMode === undefined) {
await ctx.plugin(LocalFileSystem, { cwd: root })
} else {
await ctx.plugin(SandboxPolicy, { mode: options.sandboxMode, workspaceRoot: root })
await ctx.plugin(SandboxedFileSystem, { cwd: root })
}
if (options.fsPolicy === true) await ctx.plugin(FsPolicy)
const fiber = await ctx.plugin(ToolStrReplaceEditor, config)
return { ctx, root, fiber, owner: agent(ctx, root) }
}
describe('tool-str-replace-editor', () => {
it('registers the standalone schema and configurable description', async () => {
const { ctx, fiber } = await setup({ description: 'custom editor description' })
const schema = ctx.tools.schemas()[0]
expect(ctx.tools.schemas().map(item => item.name)).toEqual(['str_replace_editor'])
expect(schema?.description).toBe('custom editor description')
const properties = (schema?.parameters as {
properties: Record<string, { type?: string; items?: { type?: string } }>
}).properties
expect(properties).not.toHaveProperty('replace_all')
expect(properties.insert_line?.type).toBe('integer')
expect(properties.view_range?.items?.type).toBe('integer')
expect(ctx.tools.get('str_replace_editor')?.presentCall?.({
command: 'view',
path: '/workspace/a.txt',
})).toMatchObject({
card: 'generic',
kind: 'read',
locations: [{ path: '/workspace/a.txt' }],
})
expect(ctx.tools.get('str_replace_editor')?.presentCall?.({
command: 'create',
path: '/workspace/a.txt',
file_text: 'hello',
})).toMatchObject({
card: 'diff',
diffs: [{ path: '/workspace/a.txt', oldText: null, newText: 'hello' }],
})
expect(ctx.tools.get('str_replace_editor')?.presentCall?.({
command: 'str_replace',
path: '/workspace/a.txt',
old_str: 'old',
new_str: 'new',
})).toMatchObject({
card: 'diff',
diffs: [{ path: '/workspace/a.txt', oldText: 'old', newText: 'new' }],
})
expect(ctx.tools.get('str_replace_editor')?.presentCall?.({
command: 'insert',
path: '/workspace/a.txt',
insert_line: 0,
new_str: 'x',
})).toMatchObject({
card: 'generic',
kind: 'edit',
locations: [{ path: '/workspace/a.txt', line: 1 }],
})
expect(ctx.tools.get('str_replace_editor')?.presentCall?.({
command: 'create',
path: '/workspace/empty.txt',
})).toMatchObject({
diffs: [{ path: '/workspace/empty.txt', oldText: null, newText: '' }],
})
expect(ctx.tools.get('str_replace_editor')?.presentCall?.({
command: 'str_replace',
path: '/workspace/a.txt',
})).toMatchObject({
diffs: [{ path: '/workspace/a.txt', oldText: null, newText: '' }],
})
expect(ctx.tools.get('str_replace_editor')?.presentCall?.({
command: 'insert',
path: '/workspace/a.txt',
})).toMatchObject({
locations: [{ path: '/workspace/a.txt' }],
})
await fiber.dispose()
expect(ctx.tools.schemas()).toEqual([])
expect(ctx.tools.get('str_replace_editor')).toBeUndefined()
})
it('creates, views, replaces, and inserts with the canonical model-facing output', async () => {
const { ctx, root, owner } = await setup()
const sample = join(root, 'sample.txt')
expect(text(await call(ctx, owner, {
command: 'create',
path: sample,
file_text: 'one\ntwo\nthree\n',
}))).toBe(`New file created successfully at: ${sample}`)
expect(text(await call(ctx, owner, {
command: 'view',
path: sample,
view_range: [2, -1],
}))).toBe([
`Here's the content of ${sample} with line numbers (which has a total of 4 lines) with view_range=[2, -1]:`,
' 2 two',
' 3 three',
' 4 ',
'',
].join('\n'))
expect(text(await call(ctx, owner, {
command: 'str_replace',
path: sample,
old_str: 'two',
new_str: 'TWO',
}))).toBe(`The file ${sample} has been edited successfully.`)
expect(text(await call(ctx, owner, {
command: 'str_replace',
path: sample,
old_str: 'TWO',
}))).toBe(`The file ${sample} has been edited successfully.`)
expect(text(await call(ctx, owner, {
command: 'insert',
path: sample,
insert_line: 1,
new_str: 'between',
}))).toBe(`The file ${sample} has been edited successfully.`)
expect(await readFile(sample, 'utf8')).toBe('one\nbetween\n\nthree\n')
})
it('writes replacement text literally', async () => {
const { ctx, root, owner } = await setup()
const sample = join(root, 'literal.txt')
const replacement = "$&|$`|$'|$$"
await writeFile(sample, 'before OLD after')
expect((await call(ctx, owner, {
command: 'str_replace',
path: sample,
old_str: 'OLD',
new_str: replacement,
})).isError).toBe(false)
expect(await readFile(sample, 'utf8')).toBe(`before ${replacement} after`)
})
it('lists visible entries to depth two and clips at the configured view limit', async () => {
const { ctx, root, owner } = await setup({ maxOutputChars: 10_000 })
await mkdir(join(root, 'dir', 'nested', 'third'), { recursive: true })
await mkdir(join(root, 'dir', 'node_modules', 'pkg'), { recursive: true })
await mkdir(join(root, 'dir', 'node_modules_old'), { recursive: true })
await mkdir(join(root, 'dir', '__pycache__'), { recursive: true })
await mkdir(join(root, 'dir', '__pycache__backup'), { recursive: true })
await writeFile(join(root, 'dir', 'visible.txt'), 'ok')
await writeFile(join(root, 'dir', '.hidden'), 'hidden')
await writeFile(join(root, 'dir', 'nested', 'child.txt'), 'child')
await writeFile(join(root, 'dir', 'nested', 'third', 'too-deep.txt'), 'deep')
await writeFile(join(root, 'dir', 'node_modules', 'pkg', 'index.js'), 'hidden dependency')
await writeFile(join(root, 'dir', 'node_modules_old', 'kept.js'), 'visible source')
await writeFile(join(root, 'dir', '__pycache__', 'module.pyc'), 'cache')
await writeFile(join(root, 'dir', '__pycache__backup', 'kept.py'), 'visible source')
const listDir = ctx.fs.listDir.bind(ctx.fs)
const otherTarget = await ctx.fs.resolve(join(root, 'dir', 'other'))
ctx.fs.listDir = async (target, signal) => {
const entries = await listDir(target, signal)
return target.displayPath === join(root, 'dir')
? [
{ name: 'same-target', type: 'other', target: otherTarget },
{ name: 'other', type: 'other', target: otherTarget },
...entries.toReversed(),
]
: entries
}
const listing = text(await call(ctx, owner, { command: 'view', path: join(root, 'dir') }))
expect(listing).not.toContain('.hidden')
expect(listing).not.toContain('too-deep.txt')
expect(listing).not.toContain('index.js')
expect(listing).not.toContain('module.pyc')
expect(listing).toContain('node_modules_old/kept.js')
expect(listing).toContain('__pycache__backup/kept.py')
const clipped = await setup({ maxOutputChars: 10 })
await writeFile(join(clipped.root, 'large.txt'), 'x'.repeat(100))
expect(text(await call(clipped.ctx, clipped.owner, {
command: 'view',
path: join(clipped.root, 'large.txt'),
})))
.toContain('<response clipped>')
})
it('matches canonical empty-line, range, and end-insert behavior', async () => {
const { ctx, root, owner } = await setup()
const empty = join(root, 'empty.txt')
const newline = join(root, 'newline.txt')
const plain = join(root, 'plain.txt')
await writeFile(empty, '')
await writeFile(newline, '\n')
await writeFile(plain, 'one\ntwo')
expect(text(await call(ctx, owner, { command: 'view', path: empty })))
.toContain('(which has a total of 1 lines):\n 1 \n')
expect(text(await call(ctx, owner, { command: 'view', path: newline })))
.toContain('(which has a total of 2 lines):\n 1 \n 2 \n')
expect(text(await call(ctx, owner, {
command: 'view',
path: plain,
view_range: [1, 2],
}))).toContain(' 2 two')
expect(text(await call(ctx, undefined, {
command: 'view',
path: plain,
}))).toContain(' 1 one')
expect((await call(ctx, undefined, {
command: 'create',
path: join(root, 'ownerless.txt'),
file_text: 'ownerless',
})).isError).toBe(false)
await call(ctx, owner, {
command: 'insert',
path: plain,
insert_line: 2,
new_str: 'three',
})
expect(await readFile(plain, 'utf8')).toBe('one\ntwo\nthree')
await writeFile(newline, 'one\n')
await call(ctx, owner, {
command: 'insert',
path: newline,
insert_line: 2,
new_str: 'three',
})
expect(await readFile(newline, 'utf8')).toBe('one\n\nthree')
})
it('uses old_str-only replacement failures and rejects relative paths', async () => {
const { ctx, root, owner } = await setup()
const ambiguous = join(root, 'ambiguous.txt')
await writeFile(ambiguous, 'same\nother\nsame')
const missing = await call(ctx, owner, {
command: 'str_replace',
path: ambiguous,
old_str: 'absent',
new_str: 'x',
})
expect(missing.isError).toBe(true)
expect(text(missing)).toContain(`old_str \`absent\` did not appear verbatim in ${ambiguous}`)
expect(text(missing)).not.toContain('old_string')
const repeated = await call(ctx, owner, {
command: 'str_replace',
path: ambiguous,
old_str: 'same',
new_str: 'x',
})
expect(repeated.isError).toBe(true)
expect(text(repeated)).toContain('Multiple occurrences of old_str `same` in lines [1, 3]')
expect(text(repeated)).not.toContain('replace_all')
await writeFile(ambiguous, 'alpha\nbeta\nmiddle\nalpha\nbeta')
const repeatedMultiline = await call(ctx, owner, {
command: 'str_replace',
path: ambiguous,
old_str: 'alpha\nbeta',
new_str: 'x',
})
expect(text(repeatedMultiline))
.toContain('Multiple occurrences of old_str `alpha\nbeta` in lines [1, 4]')
const mixedEol = join(root, 'mixed-eol.txt')
await writeFile(mixedEol, 'alpha\r\nbeta\nmiddle\nalpha\nbeta')
expect((await call(ctx, owner, {
command: 'str_replace',
path: mixedEol,
old_str: 'alpha\r\nbeta',
new_str: 'replaced',
})).isError).toBe(false)
expect(await readFile(mixedEol, 'utf8')).toBe('replaced\nmiddle\nalpha\nbeta')
const relative = await call(ctx, owner, { command: 'view', path: 'ambiguous.txt' })
expect(relative.isError).toBe(true)
expect(text(relative)).toContain('is not an absolute path')
expect(await readFile(ambiguous, 'utf8')).toBe('alpha\nbeta\nmiddle\nalpha\nbeta')
})
it('reports invalid commands or arguments without mutating files', async () => {
const { ctx, root, owner } = await setup()
const ambiguous = join(root, 'ambiguous.txt')
const empty = join(root, 'empty.txt')
const trailingNewline = join(root, 'trailing-newline.txt')
const threeLines = join(root, 'three-lines.txt')
const directory = join(root, 'directory')
await writeFile(ambiguous, 'same same')
await writeFile(empty, '')
await writeFile(trailingNewline, 'one\n')
await writeFile(threeLines, 'one\ntwo\nthree')
await mkdir(directory)
const cases = [
{ command: 'view', path: '' },
{ command: 'view', path: join(root, 'missing.txt') },
{ command: 'view', path: ambiguous, view_range: [1] },
{ command: 'view', path: ambiguous, view_range: [0, 1] },
{ command: 'view', path: ambiguous, view_range: [1.5, 2] },
{ command: 'view', path: threeLines, view_range: [1, 99] },
{ command: 'view', path: threeLines, view_range: [2, 1] },
{ command: 'view', path: directory, view_range: [1, 1] },
{ command: 'create', path: join(root, 'new.txt') },
{ command: 'create', path: ambiguous, file_text: 'overwrite' },
{ command: 'str_replace', path: ambiguous, new_str: 'x' },
{ command: 'str_replace', path: ambiguous, old_str: '', new_str: 'x' },
{ command: 'insert', path: ambiguous, new_str: 'x' },
{ command: 'insert', path: ambiguous, insert_line: -1, new_str: 'x' },
{ command: 'insert', path: ambiguous, insert_line: 1.5, new_str: 'x' },
{ command: 'insert', path: ambiguous, insert_line: 99, new_str: 'x' },
{ command: 'insert', path: empty, insert_line: 2, new_str: 'x' },
{ command: 'insert', path: directory, insert_line: 0, new_str: 'x' },
]
for (const args of cases) {
expect((await call(ctx, owner, args)).isError).toBe(true)
}
expect(await readFile(ambiguous, 'utf8')).toBe('same same')
ctx.fs.stat = async () => ({ version: FsVersion('special'), type: 'other' })
const special = await call(ctx, owner, { command: 'view', path: join(root, 'special') })
expect(special.isError).toBe(true)
expect(special.error).toMatchObject({ info: { code: 'FS_NOT_REGULAR_FILE' } })
expect((await call(ctx, owner, {
command: 'str_replace',
path: join(root, 'special'),
old_str: 'x',
new_str: 'y',
})).error).toMatchObject({ info: { code: 'FS_NOT_REGULAR_FILE' } })
expect((await call(ctx, owner, {
command: 'insert',
path: join(root, 'special'),
insert_line: 0,
new_str: 'x',
})).error).toMatchObject({ info: { code: 'FS_NOT_REGULAR_FILE' } })
})
it('delegates read-before-edit decisions to fs-policy', async () => {
const { ctx, root, owner } = await setup({}, { fsPolicy: true })
const existing = join(root, 'existing.txt')
const created = join(root, 'created.txt')
await writeFile(existing, 'before')
const blindEdit = await call(ctx, owner, {
command: 'str_replace',
path: existing,
old_str: 'before',
new_str: 'after',
})
expect(blindEdit.error).toMatchObject({ info: { code: 'FS_NOT_OBSERVED' } })
expect(await readFile(existing, 'utf8')).toBe('before')
await call(ctx, owner, { command: 'view', path: existing })
expect((await call(ctx, owner, {
command: 'str_replace',
path: existing,
old_str: 'before',
new_str: 'after',
})).isError).toBe(false)
expect(await readFile(existing, 'utf8')).toBe('after')
expect((await call(ctx, owner, {
command: 'insert',
path: existing,
insert_line: 1,
new_str: 'tail',
})).isError).toBe(false)
expect(await readFile(existing, 'utf8')).toBe('after\ntail')
expect((await call(ctx, owner, {
command: 'create',
path: created,
file_text: 'new',
})).isError).toBe(false)
expect(await readFile(created, 'utf8')).toBe('new')
})
it('passes the session sandbox policy to every mutation', async () => {
const { ctx, root, owner } = await setup({}, { sandboxMode: 'read-only' })
const path = join(root, 'blocked.txt')
const result = await call(ctx, owner, {
command: 'create',
path,
file_text: 'blocked',
})
expect(result.error).toMatchObject({ info: { code: 'FS_SANDBOX_DENIED' } })
expect(text(result)).toContain('[sandbox: file access denied under read-only mode]')
const ownerless = await call(ctx, undefined, {
command: 'create',
path: join(root, 'ownerless-blocked.txt'),
file_text: 'blocked',
})
expect(ownerless.error).toMatchObject({ info: { code: 'FS_SANDBOX_DENIED' } })
})
it('preserves tabs outside the edited region', async () => {
const { ctx, root, owner } = await setup()
const path = join(root, 'Makefile')
await writeFile(path, 'target:\n\told\nremove\n')
expect(text(await call(ctx, owner, { command: 'view', path })))
.toContain(' 2 \told')
await call(ctx, owner, {
command: 'str_replace',
path,
old_str: '\told',
new_str: '\tnew',
})
await call(ctx, owner, {
command: 'str_replace',
path,
old_str: 'remove\n',
})
await call(ctx, owner, {
command: 'insert',
path,
insert_line: 1,
new_str: '\tkept',
})
expect(await readFile(path, 'utf8')).toBe('target:\n\tkept\n\tnew\n')
})
it('reports missing sandbox-policy composition during plugin startup', async () => {
const root = await mkdtemp(join(tmpdir(), 'dsh-tool-str-replace-editor-missing-policy-'))
roots.push(root)
const ctx = new Context()
contexts.push(ctx)
await ctx.plugin(SystemPrompt)
await ctx.plugin(ToolRegistry)
await ctx.plugin(AgentRegistry)
await ctx.plugin(LocalFileSystem, { cwd: root })
Object.defineProperty(ctx.fs, 'sandboxMode', { value: 'read-only' })
await expect(ctx.plugin(ToolStrReplaceEditor))
.rejects.toThrow('the mounted filesystem confines but ctx.sandboxPolicy is missing')
})
it('maps unexpected backend write failures for replace and insert', async () => {
const { ctx, root, owner } = await setup()
const path = join(root, 'backend-error.txt')
await writeFile(path, 'old\n')
const failWrite = async (): Promise<never> => {
throw new Error('backend write failed')
}
ctx.fs.writeText = failWrite
const replace = await call(ctx, owner, {
command: 'str_replace',
path,
old_str: 'old',
new_str: 'new',
})
expect(replace.isError).toBe(true)
expect(text(replace)).toContain('backend write failed')
const insert = await call(ctx, owner, {
command: 'insert',
path,
insert_line: 1,
new_str: 'new',
})
expect(insert.isError).toBe(true)
expect(text(insert)).toContain('backend write failed')
})
it('rejects invalid plugin config', () => {
expect(() => {
ToolStrReplaceEditor.apply(new Context(), { maxOutputChars: 0 })
}).toThrow('maxOutputChars must be a positive safe integer')
expect(() => {
ToolStrReplaceEditor.apply(new Context(), { description: ' ' })
}).toThrow('description must be non-empty')
})
})

View File

@@ -0,0 +1,16 @@
{
"extends": "../../../tsconfig.base.json",
"compilerOptions": {
"rootDir": "src",
"outDir": "lib/types"
},
"include": ["src"],
"references": [
{ "path": "../../../vendor/cordis" },
{ "path": "../../core/tools" },
{ "path": "../fs" },
{ "path": "../../sandbox/sandbox" },
{ "path": "../../sandbox/sandbox-policy" },
{ "path": "../../support/invariants" }
]
}

View File

@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write packages/host/directory-picker-browse/README.md
README.md: 318380405214d5f25ad77e348c4e134a8981ffb3
README.zh.md: 2f88f64cc2974b8535e34eb9798f512ea109b754
README.md: 52b5fe7e89f915be3b50324628e9d5c48f1ef94c
README.zh.md: 742da39470083887a71ddba4a7c8012f0ce0ea1f

View File

@@ -6,7 +6,7 @@ The **in-app browsing backend** of the [directory-picker seam](../directory-pick
Behavior facts: listings return **directories only**, name-sorted, with symlinks-to-directories followed (broken/cyclic links skipped — the probe `stat` failing means "not enterable") and a host-owned `hidden` flag (POSIX dot convention) left for the client to act on; `crumbs` is the root-to-target ancestor chain, the root crumb labeled by its full path (`/`, `C:\`); an absent `list` path means the host account's home directory. `createDirectory` is non-recursive (a missing parent is a real failure, not a level to invent) and validates the name as a single non-blank segment even when called directly, mirroring the wire schema's fence. Both primitives reject an explicit path that is not fully qualified — relative forms, and on Windows the rooted drive-less forms (`\foo`, `/foo`) and incomplete UNC prefixes (`\\`, `\\server`) that `isAbsolute` accepts — with `directory-unreadable`/`directory-create-failed`, instead of letting `resolve` rebase it under the host process cwd or current drive. One `list` call returns at most `maxEntries` rows (config, default 1000 — the bound GitHub's web UI applies to directory listings), and the level streams through a bounded window so memory stays O(maxEntries) no matter how many children the directory holds: a cut level keeps the name-sorted head, counts hidden rows against the bound, probes only windowed candidates, and reports `truncated: true` so the client can say the level is incomplete (a windowed broken symlink is not backfilled from beyond the window — the eviction already marks the level truncated); window insertion is binary with an O(1) full-window tail rejection, and `list` threads the caller's `AbortSignal` so a disconnect or timeout stops the scan instead of letting it outlive the caller. Failures throw the seam's typed `DirectoryPickerError`. Policy rationale: [the directory-picker capability seam Agent Note](../../../.agents/notes/implemented/architecture/2026-07-28-directory-picker-capability-seam.md).
**Dual-face package**: the browser half (`./client`) fills [ui-workspace's](../../client/ui-workspace/README.md) two directory-flow holes with the in-app **Select Workspace Directory** dialog (figma `Harness` 813-23126 family — Miller two-column view, breadcrumb with a click-to-edit path zone, nested New-folder dialog), driving `host.listDirectory`/`host.createDirectory` and registering its own locale namespace (`directory-browser`, zh default / en). One cordis.yml row therefore composes both sides of the browse interaction; the client carries no capability-kind branching, and mounting a second flow package fails at load (the holes are `single` kind).
**Dual-face package**: the browser half (`./client`) fills [ui-workspace's](../../client/ui-workspace/README.md) two directory-flow holes with the in-app **Select Workspace Directory** dialog (figma `Harness` 813-23126 family — Miller two-column view whose navigations land selection-anchored and quiet: the previous view keeps rendering while a crumb jump or a submitted path is scanned (a "Loading…" pill floats over it only once the scan outlives a 300ms silence window, never shifting the columns), then target and parent legs land as one two-pane frame with the target re-selected as its actual parent-level entry — so stepping back never collapses and no intermediate frame flashes (a parent leg outliving its 200ms wait bound lands the target alone and upgrades in place; a failed or truncated parent leg keeps the single-pane landing; the display root keeps the single wide level); breadcrumb with a click-to-edit path zone whose editor seeds a trailing separator, prefix-filters the listed level from the draft's final segment while typing (case-insensitively, over the listed — possibly truncated — rows only; Enter still navigates by the exact text), and cancels on Escape or when focus leaves the dialog card (window/tab switches and in-card focus moves keep the draft); a fixed-label show-hidden footer toggle over the host's `hidden` flags, with a dot-led typed prefix revealing its matches and the current selection exempt from both filters; nested New-folder dialog), driving `host.listDirectory`/`host.createDirectory` and registering its own locale namespace (`directory-browser`, zh default / en). One cordis.yml row therefore composes both sides of the browse interaction; the client carries no capability-kind branching, and mounting a second flow package fails at load (the holes are `single` kind).
## Model Experience

View File

@@ -6,7 +6,7 @@
行为事实:列举**只返回目录**、按名称排序,指向目录的符号链接会被跟随(断链/循环链接被跳过——探测 `stat` 失败即"不可进入"),并携带宿主判定的 `hidden` 标志POSIX 点前缀约定),展示决策留给客户端;`crumbs` 是从根到目标的祖先链,根 crumb 以完整路径标注(`/``C:\``list` 不带路径即列举宿主账户的家目录。`createDirectory` 不递归(父目录缺失是真实失败,不是要补造的层级),且即便被直接调用也把名称校验为单个非空段,与协议 schema 的栅栏一致。两个原语都拒绝非完全限定的显式路径——相对形态,以及 Windows 上 `isAbsolute` 会放行的无盘符有根形态(`\foo``/foo`)与不完整的 UNC 前缀(`\\``\\server`)——报 `directory-unreadable``directory-create-failed`,而不是任由 `resolve` 把它重定位到宿主进程 cwd 或当前盘符之下。单次 `list` 至多返回 `maxEntries` 行(配置项,默认 1000——GitHub 网页端对目录列举采用的同一上限),且层级以流式方式经过一个有界窗口,无论目录有多少子项内存都保持 O(maxEntries):被截断的层级保留按名排序的头部、隐藏行计入上限、只探测窗口内候选,并报告 `truncated: true`,供客户端提示层级不完整(窗口内的断链符号链接不会从窗口外回填——发生过驱逐本身已把层级标记为截断);窗口插入为二分查找、满窗尾部单次比较即拒绝,且 `list` 透传调用方的 `AbortSignal`,断连或超时会停止扫描而不是让它在调用方离开后继续。失败抛出 seam 的类型化 `DirectoryPickerError`。策略依据:[目录选择能力 seam Agent Note](../../../.agents/notes/implemented/architecture/2026-07-28-directory-picker-capability-seam.md)。
**双面包**browser half`./client`)以应用内 **选择工作区目录** 对话框figma `Harness` 813-23126 家族——Miller 双列视图、带点击即编辑路径区的面包屑、嵌套新建文件夹对话框)填入 [ui-workspace](../../client/ui-workspace/README.md) 的两个目录流洞,驱动 `host.listDirectory``host.createDirectory`,并注册自己的 locale 命名空间(`directory-browser`zh 默认en。因此一行 cordis.yml 同时组合浏览交互的两侧client 侧不含任何能力 kind 分支,挂载第二个流程包会在加载期失败(洞为 `single` kind
**双面包**browser half`./client`)以应用内 **选择工作区目录** 对话框figma `Harness` 813-23126 家族——Miller 双列视图,其导航以选中项为锚、安静落地:扫描 crumb 跳转或提交的路径期间,先前视图持续渲染("Loading…" 胶囊仅在扫描超出 300ms 静默窗口后才浮于其上,绝不挪动各列),随后目标与父层级两程以单个双栏帧落地,目标被重新选中为其在父层级中的实际条目——因此后退绝不塌缩,也没有中间帧闪现(父层级这一程超出其 200ms 等待上限时目标单独落地随后就地升级父层级这一程失败或被截断时保持单栏落地展示根保持单个宽层级带点击即编辑路径区的面包屑其编辑器预填尾随分隔符、输入时以草稿末段对所列层级做前缀过滤不区分大小写且仅作用于已列出、可能被截断的行Enter 仍按确切文本导航)、按 Escape 或焦点离开对话框卡片即取消(窗口/标签页切换与卡片内焦点移动保留草稿);基于宿主 `hidden` 标志、标签固定的"显示隐藏"footer 开关,键入以点开头的前缀会显出其匹配项,且当前选中项不受这两种过滤影响;嵌套新建文件夹对话框)填入 [ui-workspace](../../client/ui-workspace/README.md) 的两个目录流洞,驱动 `host.listDirectory``host.createDirectory`,并注册自己的 locale 命名空间(`directory-browser`zh 默认en。因此一行 cordis.yml 同时组合浏览交互的两侧client 侧不含任何能力 kind 分支,挂载第二个流程包会在加载期失败(洞为 `single` kind
## 模型体验

View File

@@ -1,6 +1,7 @@
/* Directory-browser dialog (figma 813-23126 family). The shared Modal renders
* headless here — mask, card, Escape only — and this module owns the figma
* frame: 600×420 card (viewport-clamped), header (title + crumbs, l3 separator),
* frame: 680×500 card (viewport-clamped; upsized from the figma 600×420),
* header (title + crumbs, l3 separator),
* the one-or-two-column Miller content, and the bordered footer. */
/* Doubled class beats Modal's own .dialog regardless of stylesheet order. */
@@ -8,19 +9,32 @@
* columns scroll, so shrinking the height keeps Open/Cancel reachable
* instead of clipping them below a fixed overlay. */
.dialog.dialog {
width: min(600px, 100%);
height: min(420px, calc(100dvh - 32px));
width: min(680px, 100%);
height: min(500px, calc(100dvh - 32px));
padding: 0;
gap: 0;
/* The Modal card is an l2 surface and the columns below scroll on it:
* rebind the scrollbar indirection to the elevation pair here, on the
* surface, so it inherits down to whichever descendant scrolls (the
* rebinding contract in ui-theme styles/scrollbar.css). */
--dsh-scrollbar-thumb: var(--dsw-alias-scrollbar-bg-l2);
--dsh-scrollbar-thumb-hover: var(--dsw-alias-scrollbar-hover-l2);
}
/* Header block: pl24 pr14 pt22 pb12, 8px between title row and crumb row. */
/* Card-scope wrapper hosting the path editor's Escape and focus-leave
* observers; display:contents keeps header/content/footer as direct flex
* children of the Modal card. */
.editorScope {
display: contents;
}
/* Header block: pl24 pr14 pt16 pb8, 8px between title row and crumb row. */
.header {
display: flex;
flex-direction: column;
gap: 8px;
flex: none;
padding: 22px 14px 12px 24px;
padding: 16px 14px 8px 24px;
border-bottom: 1px solid var(--dsw-alias-border-l3);
}
@@ -54,8 +68,12 @@
align-items: stretch;
flex: 1 1 0;
min-height: 0;
gap: 20px;
/* 12px of row gap on each side of the divider; the left side reads wider
* by the column's trailing 8px scrollbar clearance, which is deliberate —
* the thumb needs that room, the right pane's rows do not. */
gap: 12px;
overflow-x: auto;
scrollbar-width: none;
}
.crumbTrail {
@@ -126,28 +144,36 @@
color: var(--dsw-alias-label-primary);
}
/* Miller content: pt16 px24; columns are 256 wide (or full width solo) with
* the hairline divider centered between them; each column scrolls alone. */
/* Miller content: symmetric 16px vertical padding so the divider clears the
* header and footer rules evenly; each column scrolls alone (column widths
* live at .column). */
.content {
display: flex;
flex-direction: column;
flex: 1 1 0;
min-height: 0;
padding: 16px 24px 0;
/* Anchors the floating loading pill (.loadingFloat). */
position: relative;
/* Right inset is slimmer than the left: the trailing column's own 8px
* scrollbar clearance makes up the optical difference. */
padding: 16px 16px 16px 24px;
}
/* Columns split the row evenly around the divider (a solo column takes the
* whole row); 256px is the floor below which the row scrolls (scrollbar
* hidden, the effect pins the child pane into view) instead of squeezing
* the panes. */
.column {
display: flex;
flex-direction: column;
gap: 2px;
width: 256px;
flex: none;
overflow-y: auto;
}
.columnWide {
width: 100%;
flex: 1 1 0;
min-width: 256px;
overflow-y: auto;
/* The themed scrollbar occupies the column's edge (styled scrollbars are
* classic, gutter-taking ones); the extra clearance keeps the row pills
* clear of the thumb. */
padding-right: 8px;
}
.divider {
@@ -216,6 +242,10 @@
.status,
.error {
padding: 4px;
/* The loading pill occupies the opposite corner while a stale status stays
* visible. Reserve its widest localized footprint so wrapped text cannot
* run underneath it on a narrow card. */
padding-right: 120px;
font-size: 12px;
line-height: 18px;
}
@@ -228,8 +258,25 @@
color: var(--dsw-alias-state-error-primary);
}
/* Footer: l3 separator on top, pt12 px24, New-folder pinned left; the fixed
* card leaves the figma 28px below the 36px buttons. */
/* The slow-scan indicator floats over the content's bottom-RIGHT corner on
* the card background instead of occupying a row: a scan must never shift
* the columns' height, and the stale view keeps rendering beneath it (it
* only appears at all once a scan outlives SLOW_SCAN_DELAY_MS). Right,
* not left: the truncated/error status rows flow at the bottom LEFT and
* stay on screen through a scan, with their reserved right padding keeping
* both legible even on a narrow card. After .status in the cascade — the
* element carries both classes and this padding must win the
* same-specificity race. */
.loadingFloat {
position: absolute;
right: 16px;
bottom: 8px;
padding: 2px 8px;
background: var(--dsw-alias-bg-layer-2);
}
/* Footer: l3 separator on top, symmetric padding so the row sits vertically
* centered in the bar; New-folder and the show-hidden toggle pin left. */
.footerBar {
display: flex;
align-items: center;
@@ -238,10 +285,42 @@
flex-wrap: wrap;
gap: 8px;
flex: none;
padding: 12px 24px 28px;
padding: 16px 24px;
border-top: 1px solid var(--dsw-alias-border-l3);
}
/* Show-hidden toggle: a subtle fixed-label text button left of the gap;
* the pressed state seats a check glyph after the label (Menu's selected
* vocabulary; trailing so the label never shifts) instead of flipping the
* wording. */
.showHiddenToggle {
display: inline-flex;
align-items: center;
gap: 4px;
border: none;
background: transparent;
padding: 0;
font-size: 13px;
line-height: 20px;
font-weight: 500;
color: var(--dsw-alias-label-secondary);
cursor: pointer;
white-space: nowrap;
}
.showHiddenToggle:hover {
color: var(--dsw-alias-label-primary);
}
.showHiddenToggle:disabled {
color: var(--dsw-alias-label-caption);
cursor: default;
}
.showHiddenToggleActive {
color: var(--dsw-alias-label-primary);
}
.footerGap {
flex: 1 1 0;
}

View File

@@ -1,22 +1,32 @@
/**
* The in-app workspace-directory browser (figma Harness 813-23126 family): a
* 600×420 dialog (clamped to short/narrow viewports — the Miller row scrolls
* 680×500 dialog (clamped to short/narrow viewports — the Miller row scrolls
* sideways, the columns scroll down) whose header carries the title, the selection-path
* breadcrumb, and a click-to-edit path zone; below it a Miller view — one
* full-width level until a row is selected, then two 256px columns (level |
* selected folder's children) around a hairline divider. Selecting in the
* full-width level until a row is selected, then two columns splitting the
* row evenly (256px floor; level | selected folder's children) around a
* hairline divider. Navigations land selection-anchored and quiet: the
* previous view keeps rendering while a crumb jump or a submitted path is
* scanned, then target and parent legs land as one two-pane frame (a slow
* parent leg falls back to landing the target alone and upgrading in
* place), so stepping back keeps two panes away from the display root and
* navigation never flashes an intermediate frame. Selecting in the
* right column shifts the view one level deeper. "New folder" opens a nested
* create dialog targeting the selected folder (or the level itself) and
* selects the created folder. Open adopts the selected folder, falling back
* to the listed level. Pure consumer of the injected browse calls — the
* owning flow decides what "Open" means and owns the workspace-creation
* error surface. Hidden entries are host-flagged and filtered here (a
* show-hidden toggle is deferred work, client-side only).
* error surface. Hidden entries are host-flagged and hidden by default; the
* footer's fixed-label "Show hidden files" toggle (aria-pressed, check when
* on) reveals them (client-side only). The path editor opens seeded with a
* trailing separator, and while the draft's directory part names a listed
* level, its final segment prefix-filters that level's rows (a dot-led
* prefix also reveals the hidden entries it names).
*/
import { useCallback, useEffect, useRef, useState } from 'react'
import clsx from 'clsx'
import {
Button, IconChevronRightOutline14, IconFolderClose16, IconFolderOpen16, IconPlusOutline16, Modal,
Button, IconCheckOutline16, IconChevronRightOutline14, IconFolderClose16, IconFolderOpen16, IconPlusOutline16, Modal,
} from '@deepseek-ai/dsh-client-ui-primitives'
import type { DirectoryEntry, DirectoryListing } from '@deepseek-ai/dsh-client-runtime/client'
import { DirectoryBrowseError } from '@deepseek-ai/dsh-client-runtime/client'
@@ -47,6 +57,24 @@ function failureText(error: unknown): string {
return error instanceof Error ? error.message : String(error)
}
/**
* How long a scan may stay visually silent before the floating "Loading…"
* pill appears. The stale view keeps rendering while a scan is in flight, so
* a listing that settles inside this window swaps the panes with no
* intermediate frame at all; only a genuinely slow host (a network mount, a
* cold disk) surfaces the indicator.
*/
const SLOW_SCAN_DELAY_MS = 300
/**
* How long a navigation landing waits for its parent leg before committing
* the target alone. Inside the window both legs land as ONE two-pane frame —
* no single-pane flash between them; past it the target commits single-pane
* at once (an Enter-submitted navigation is never held hostage by a stalled
* parent) and the late parent leg upgrades the landing in place.
*/
const PARENT_LEG_WAIT_MS = 200
/**
* Breadcrumb rows for display: inside the home subtree the chain starts at a
* localized Home crumb; outside it the full ancestry shows, the root labeled
@@ -59,17 +87,59 @@ function displayCrumbs(listing: DirectoryListing, homeLabel: string): DirectoryE
return [{ name: homeLabel, path: listing.home, hidden: false }, ...tail]
}
/**
* The listing's platform separator, inferred from the home path the host
* stamped — never from typed text or entry paths, where a backslash is a
* legal POSIX name character. Still a heuristic at the last step: a POSIX
* home directory whose own name contains a backslash would misread.
* TODO: replace with a host-stamped `separator` field on the wire
* DirectoryListing so the platform fact travels verbatim (the trade-off is
* recorded in the directory-picker capability seam Agent Note).
*/
function separatorOf(listing: DirectoryListing): '\\' | '/' {
return listing.home.includes('\\') ? '\\' : '/'
}
/**
* The path draft's final segment, when its directory part is exactly the
* level `listing` lists — the segment the level prefix-filters on while the
* user types. Any other draft (no separator yet, or naming some other
* directory) leaves the level unfiltered. The directory part compares
* exactly (it is the host's own path text, reached by seeding or erasing);
* only the name filter downstream is case-insensitive.
*/
function draftPrefixFor(listing: DirectoryListing, draft: string | null): string | null {
if (draft === null) return null
const sep = separatorOf(listing)
const cut = draft.lastIndexOf(sep)
if (cut === -1) return null
const level = listing.path.endsWith(sep) ? listing.path : `${listing.path}${sep}`
return draft.slice(0, cut + 1) === level ? draft.slice(cut + 1) : null
}
/** One column of folder rows (the Miller view renders one or two of these). */
function LevelColumn({ entries, selectedPath, busy, onPick, wide }: {
function LevelColumn({ entries, selectedPath, busy, onPick, showHidden, filterPrefix, pathEditing }: {
entries: readonly DirectoryEntry[]
selectedPath: string | null
busy: boolean
onPick: (entry: DirectoryEntry) => void
wide: boolean
showHidden: boolean
filterPrefix: string | null
pathEditing: boolean
}) {
const visible = entries.filter((entry) => {
// The selection is exempt from both filters: it anchors the two-pane
// view (crumbs and the child pane point at it), so neither the hidden
// filter after a dot-reveal pick nor a prefix miss may orphan it.
if (entry.path === selectedPath) return true
if (filterPrefix !== null && !entry.name.toLowerCase().startsWith(filterPrefix.toLowerCase())) return false
// A dot-led prefix names hidden entries explicitly, so matching ones
// surface even while the toggle keeps the rest hidden.
return showHidden || !entry.hidden || filterPrefix?.startsWith('.') === true
})
return (
<div className={clsx(css.column, wide && css.columnWide)} role="list">
{entries.filter(entry => !entry.hidden).map((entry) => {
<div className={css.column} role="list">
{visible.map((entry) => {
const selected = entry.path === selectedPath
return (
// The wrapper carries the list semantics; the row keeps its NATIVE
@@ -80,6 +150,15 @@ function LevelColumn({ entries, selectedPath, busy, onPick, wide }: {
aria-current={selected || undefined}
className={clsx(css.row, selected && css.rowSelected)}
disabled={busy}
// While the path editor is open, keep focus in it: a focus
// steal on mousedown would blur the editor and (in engines
// where the blur lands before our guards) drop this click.
// Outside editing, rows keep native focus behavior.
onMouseDown={pathEditing ? (event) => { event.preventDefault() } : undefined}
// Editing-time focus parking happens after commit (the
// DirectoryBrowser refocus effect): a right-pane pick replaces
// this very column, so focusing the clicked node here would
// still fall to body.
onClick={() => { onPick(entry) }}
>
{selected
@@ -107,9 +186,19 @@ export function DirectoryBrowser({ open, listDirectory, createDirectory, onOpen,
const [selected, setSelected] = useState<DirectoryEntry | null>(null)
const [child, setChild] = useState<DirectoryListing | null>(null)
const [loading, setLoading] = useState(false)
// Derived from `loading` and `scanWindow` by the slow-scan effect below:
// true only once the current listing call has been in flight for
// SLOW_SCAN_DELAY_MS, so fast listings never render the indicator at all.
const [slowScan, setSlowScan] = useState(false)
// Every listing call owns a fresh silence window. `loading` may stay true
// across a superseding row pick or across a navigation's target and parent
// legs, so its boolean edge cannot identify the start of each scan.
const [scanWindow, setScanWindow] = useState(0)
const [error, setError] = useState<string | null>(null)
// Path-edit state: null = breadcrumb mode; a string = the draft being typed.
const [pathDraft, setPathDraft] = useState<string | null>(null)
// Show-hidden toggle state (pure client-side filter, reset on each open).
const [showHidden, setShowHidden] = useState(false)
// Create-folder state: null = closed; a string = the nested dialog's draft.
const [folderDraft, setFolderDraft] = useState<string | null>(null)
const [creatingFolder, setCreatingFolder] = useState(false)
@@ -148,36 +237,128 @@ export function DirectoryBrowser({ open, listDirectory, createDirectory, onOpen,
return ++requestSeq.current
}, [])
/** Hide any prior indicator and start a fresh silence window for one listing call. */
const restartSlowScanWindow = useCallback((): void => {
setSlowScan(false)
setScanWindow(value => value + 1)
}, [])
/** Launch one listing under a fresh controller so a later supersession can abort it. */
const launchListing = useCallback((path: string | undefined): { seq: number; scan: Promise<DirectoryListing> } => {
const seq = supersede()
const controller = new AbortController()
scanController.current = controller
restartSlowScanWindow()
return { seq, scan: listDirectory(path, controller.signal) }
}, [supersede, listDirectory])
}, [supersede, restartSlowScanWindow, listDirectory])
/** Replace the whole view with one freshly listed level (no selection). */
/**
* Launch a follow-up listing under the CURRENT supersession seq: a newer
* intent aborts it like the leg it continues, and it supersedes nothing.
*/
const continueScan = useCallback((path: string): Promise<DirectoryListing> => {
const controller = new AbortController()
scanController.current = controller
restartSlowScanWindow()
return listDirectory(path, controller.signal)
}, [restartSlowScanWindow, listDirectory])
/**
* Replace the whole view with a freshly navigated level. Away from the
* display root — the same collapse the crumb header renders, so crumbs and
* pane shape never disagree — the landing is two-pane: the target's ACTUAL
* parent-level entry re-selected (left pane = parent, right pane = the
* target), so a crumb jump reads as stepping back one pane. Both legs land
* as one frame when the parent leg settles within
* {@link PARENT_LEG_WAIT_MS}; past that bound (or at the display root) the
* target commits alone — single wide level, the editor closes, loading
* ends — and a late parent leg still upgrades the landing in place. A
* failed parent leg, or a truncated parent window that lacks the target,
* leaves the single-pane landing — the upgrade must never orphan the
* selection it exists to anchor. Until whichever commit comes first, the
* previous view keeps rendering: navigation swaps the panes, it never
* blanks them.
*/
const navigate = useCallback((path?: string) => {
const { seq, scan } = launchListing(path)
setLoading(true)
setError(null)
scan.then((next) => {
scan.then((target) => {
if (seq !== requestSeq.current) return
setParent(next)
setSelected(null)
setChild(null)
setLoading(false)
setPathDraft(null)
// The single-pane landing; `landed` makes it first-commit-only, while
// the two-pane commit below may still upgrade an already-landed view.
let landed = false
const landSingle = (): void => {
if (landed || seq !== requestSeq.current) return
landed = true
setParent(target)
setSelected(null)
setChild(null)
setLoading(false)
setPathDraft(null)
}
// Arity is label-independent: only the collapsed chain's depth decides.
if (displayCrumbs(target, '').length < 2) { landSingle(); return }
const parentCrumb = target.crumbs.at(-2)
/* v8 ignore next -- narrowing: a two-deep display chain implies a parent crumb (root-to-target inclusive). */
if (parentCrumb === undefined) { landSingle(); return }
continueScan(parentCrumb.path).then((parentLevel) => {
if (seq !== requestSeq.current) return
// Windows resolves a typed path preserving its case; anchor on the
// parent level's actual entry so selection comparisons hold.
const sep = separatorOf(parentLevel)
const fold = (value: string): string => (sep === '\\' ? value.toLowerCase() : value)
const match = parentLevel.entries.find(entry => fold(entry.path) === fold(target.path))
if (match === undefined) { landSingle(); return }
landed = true
setParent(parentLevel)
setSelected(match)
setChild(target)
// Idempotent on a late upgrade of a timed-out landing: reopening the
// editor or starting a newer scan supersedes this seq, so reaching
// here means the draft is closed and the loading flag is this
// navigation's own.
setLoading(false)
setPathDraft(null)
}, () => {
// The parent-leg failure (its abort included) never surfaces: the
// target listed fine, and nobody asked to see the parent level.
landSingle()
})
window.setTimeout(landSingle, PARENT_LEG_WAIT_MS)
}, (reason: unknown) => {
if (seq !== requestSeq.current) return
setLoading(false)
setError(failureText(reason))
})
}, [launchListing])
}, [launchListing, continueScan])
/** Select a row of the listed level and preview its children on the right. */
// Editor-close focus parking (consumed by the refocus effect below the
// miller-row ref): a pick parks on the selection's row, Enter and an
// input-focused Escape park on the crumb edit zone that replaces the
// input. Pointer-out cancels never set (or clear) these — yanking focus
// back from wherever the user clicked would be worse than the fall.
const refocusPick = useRef(false)
const refocusEditZone = useRef(false)
const pathInputRef = useRef<HTMLInputElement | null>(null)
const editZoneRef = useRef<HTMLButtonElement | null>(null)
/**
* Select a row of the listed level and preview its children on the right.
* Deliberately NOT one-frame like navigate(): a pick's first duty is the
* immediate selected state on the clicked row, and the pane split IS that
* feedback (aria-current pill, crumbs following the selection) — holding
* it back for the child listing would make clicks feel dropped. The quiet
* rule governs whole-view replacement, where nothing acknowledges the
* click but the swap itself.
*/
const select = useCallback((entry: DirectoryEntry) => {
const { seq, scan } = launchListing(entry.path)
// A pick while the path editor is open adopts the (filtered) row and
// closes the editor — the draft served its purpose. Focus re-parks on
// the selection after commit (see the refocus effect below).
if (pathDraft !== null) refocusPick.current = true
setPathDraft(null)
setSelected(entry)
setChild(null)
setLoading(true)
@@ -193,8 +374,31 @@ export function DirectoryBrowser({ open, listDirectory, createDirectory, onOpen,
// An unreadable selection cannot be the committing target while the
// breadcrumb still names the level: fall back to the single pane.
setSelected(null)
// Clearing the selection can unmount the very row the pick parked
// focus on (a dot-revealed hidden row re-hides); the refocus effect
// re-parks on the edit zone only if focus actually fell to body.
refocusEditZone.current = true
})
}, [launchListing])
}, [launchListing, pathDraft])
/** Abandon path editing (Escape or clicking away) and restore the crumb view. */
const cancelPathEdit = useCallback(() => {
// Cancel also withdraws a navigation the editor already launched: its
// late success must not jump to the cancelled path, so the pending
// request is superseded and the view leaves the loading state.
supersede()
setLoading(false)
setPathDraft(null)
setError(null)
// Editing may have superseded the selection's preview request; a
// selection with no preview would render a half-empty two-pane view, so
// cancel falls back to the single-pane level.
if (child === null) setSelected(null)
// With no level listed yet (the editor superseded the initial home
// listing), plain cancellation would leave a permanently blank picker:
// restart the home listing.
if (parent === null) navigate()
}, [supersede, child, parent, navigate])
/** A right-column pick advances the view one level: child becomes the level. */
const advance = useCallback((entry: DirectoryEntry) => {
@@ -213,14 +417,24 @@ export function DirectoryBrowser({ open, listDirectory, createDirectory, onOpen,
setSelected(null)
setChild(null)
setCreatingFolder(false)
setShowHidden(false)
navigate()
return
}
supersede()
// Closing mid-scan leaves nothing to load: without this edge the
// slow-scan effect keeps arming while hidden and the reopened dialog
// would show the indicator on its first frame instead of waiting out a
// fresh silence window (reopen's navigate() produces no loading edge).
setLoading(false)
setError(null)
setPathDraft(null)
setFolderDraft(null)
setCreateError(null)
// A close mid-flight (failed Enter, then Cancel) may leave refocus
// flags armed; retire them so a later render cannot consume them.
refocusPick.current = false
refocusEditZone.current = false
}, [open, navigate, supersede])
/** The folder a create or Open acts on: the selection, else the listed level. */
@@ -249,6 +463,10 @@ export function DirectoryBrowser({ open, listDirectory, createDirectory, onOpen,
// create target becomes the listed level and the new folder its selection.
const { seq, scan } = launchListing(targetPath)
setLoading(true)
// Symmetric with navigate/select: a launched scan clears the stale
// failure text (and keeps the floating indicator's corner the only
// occupant of the content's right edge while it shows).
setError(null)
scan.then((level) => {
/* v8 ignore next -- same fence as navigate/select; the modal blocks superseding input */
if (seq !== requestSeq.current) return
@@ -268,6 +486,19 @@ export function DirectoryBrowser({ open, listDirectory, createDirectory, onOpen,
})
}
// The slow-scan gate for the loading indicator: each listing call restarts
// the timer even when a superseding scan or a navigation's parent leg keeps
// `loading` continuously true. A settle inside its own window means the swap
// happened with nothing shown.
useEffect(() => {
if (!loading) {
setSlowScan(false)
return
}
const timer = window.setTimeout(() => { setSlowScan(true) }, SLOW_SCAN_DELAY_MS)
return () => { window.clearTimeout(timer) }
}, [loading, scanWindow])
// After the hooks: a closed dialog renders nothing and evaluates no copy.
const crumbSource = child ?? parent
const crumbs = crumbSource === null ? [] : displayCrumbs(crumbSource, t('browser.home'))
@@ -285,6 +516,37 @@ export function DirectoryBrowser({ open, listDirectory, createDirectory, onOpen,
const row = millerRowRef.current
if (row !== null && childPath !== undefined) row.scrollLeft = row.scrollWidth
}, [childPath])
// Every editor exit that would drop focus to body re-parks it after
// commit, so keyboard traversal stays inside the dialog (the Modal has no
// focus trap): a pick lands on the selection's row — aria-current in the
// freshly rendered left pane, which survives even a right-pane advance
// replacing the picked button's column — while Enter and an input-focused
// Escape land on the crumb edit zone that replaces the input.
useEffect(() => {
if (pathDraft !== null) return
if (refocusPick.current) {
refocusPick.current = false
refocusEditZone.current = false
const rowHost = millerRowRef.current
/* v8 ignore next -- narrowing guard: the miller row is mounted whenever a pick just committed. */
if (rowHost === null) return
const row = rowHost.querySelector<HTMLButtonElement>('button[aria-current="true"]')
/* v8 ignore next -- narrowing guard: the pick that set the flag just rendered its aria-current row. */
if (row === null) return
row.focus()
return
}
if (refocusEditZone.current) {
refocusEditZone.current = false
// Re-park only when the close actually dropped focus to body; focus
// the user parked elsewhere (a surviving row) stays theirs.
if (document.activeElement !== document.body) return
const zone = editZoneRef.current
/* v8 ignore next -- narrowing guard: crumb mode renders the edit zone whenever the editor just closed. */
if (zone === null) return
zone.focus()
}
})
if (!open) return null
const twoPane = selected !== null
@@ -310,149 +572,220 @@ export function DirectoryBrowser({ open, listDirectory, createDirectory, onOpen,
className={clsx(css.dialog)}
headless
>
<div className={css.header}>
<h2 className={css.title}>{t('browser.title')}</h2>
<div className={css.crumbBar}>
{pathDraft === null
? (
<>
<span className={css.crumbTrail} role="navigation" ref={crumbTrailRef}>
{crumbs.map((crumb, index) => (
<span key={crumb.path} className={css.crumbSeat}>
{index > 0 && <IconChevronRightOutline14 size={12} className={css.crumbChevron} />}
<button
type="button"
className={css.crumb}
disabled={parentInert}
onClick={() => { navigate(crumb.path) }}
>
{crumb.name}
</button>
</span>
))}
</span>
{/* The empty zone right of the crumbs is the path-edit affordance. */}
<button
type="button"
className={css.crumbEditZone}
aria-label={t('browser.editPath')}
// Stays available with no listed level: when the home
// listing itself fails, typing an absolute path is the one
// remaining way forward.
disabled={parentInert}
onClick={() => {
{/* Path-edit cancellation is observed at the card scope, not the
* input: once Tab parks focus on a filtered row the input is off the
* event path, yet Escape must still collapse the editor (not the
* dialog) and a further focus move out of the card must still
* cancel. display:contents keeps header/content/footer as direct
* flex children of the Modal card. */}
<div
className={css.editorScope}
onKeyDown={(event) => {
if (event.key !== 'Escape' || pathDraft === null) return
// stopPropagation keeps the card-scope Escape from the Modal's
// document listener — the same containment the input previously
// provided for itself.
event.stopPropagation()
// Escape while the input holds focus is about to unmount it; with
// focus already parked on a row, that row survives the cancel and
// keeps focus naturally. Assignment (not a conditional set) also
// retires a stale flag a failed or still-upgrading Enter left.
refocusEditZone.current = document.activeElement === pathInputRef.current
cancelPathEdit()
}}
// Focus leaving THIS dialog card while editing cancels like Escape.
// Guarded non-cancel paths: window/tab focus loss (document no
// longer focused); a focus move that stays inside the card (Tab
// onto the filtered rows or the footer toggle); and pointer paths,
// where rows and the toggle suppress focus steal on mousedown while
// editing so their click lands first. Enter keeps focus in the
// input while its navigation is in flight, so a submitted path is
// never withdrawn here. Anchored to this card via closest, not any
// [role="dialog"], so focus escaping into a sibling overlay cancels.
onBlur={(event) => {
if (pathDraft === null) return
if (!document.hasFocus()) return
const card = event.currentTarget.closest('[role="dialog"]')
/* v8 ignore next -- narrowing guard: this scope always renders inside the Modal card. */
if (card === null) return
if (event.relatedTarget instanceof Node && card.contains(event.relatedTarget)) return
// The user moved focus out of the card themselves: cancel without
// re-parking (a lingering Enter-failure flag must not yank focus
// back either).
refocusEditZone.current = false
cancelPathEdit()
}}
>
<div className={css.header}>
<h2 className={css.title}>{t('browser.title')}</h2>
<div className={css.crumbBar}>
{pathDraft === null
? (
<>
<span className={css.crumbTrail} role="navigation" ref={crumbTrailRef}>
{crumbs.map((crumb, index) => (
<span key={crumb.path} className={css.crumbSeat}>
{index > 0 && <IconChevronRightOutline14 size={12} className={css.crumbChevron} />}
<button
type="button"
className={css.crumb}
disabled={parentInert}
onClick={() => { navigate(crumb.path) }}
>
{crumb.name}
</button>
</span>
))}
</span>
{/* The empty zone right of the crumbs is the path-edit affordance. */}
<button
type="button"
className={css.crumbEditZone}
aria-label={t('browser.editPath')}
// Stays available with no listed level: when the home
// listing itself fails, typing an absolute path is the one
// remaining way forward.
disabled={parentInert}
ref={editZoneRef}
onClick={() => {
// Opening the editor supersedes any pending listing: a
// settlement landing before the first keystroke would
// otherwise close the editor via navigate's draft reset.
supersede()
setLoading(false)
setPathDraft(selected?.path ?? parent?.path ?? '')
}}
/>
</>
)
: (
<input
className={css.pathInput}
value={pathDraft}
aria-label={t('browser.editPath')}
autoFocus
disabled={parentInert}
onChange={(event) => {
supersede()
setLoading(false)
// Seed with a trailing separator so typing immediately
// continues into child names (and prefix-filters below).
// No listed level means nothing to seed from (the editor
// is the recovery path for a failed home listing).
if (parent === null) {
setPathDraft('')
return
}
const base = selected?.path ?? parent.path
const sep = separatorOf(parent)
setPathDraft(base.endsWith(sep) ? base : `${base}${sep}`)
}}
/>
</>
)
: (
<input
className={css.pathInput}
value={pathDraft}
aria-label={t('browser.editPath')}
autoFocus
ref={pathInputRef}
disabled={parentInert}
onChange={(event) => {
// Editing the draft supersedes any in-flight navigation:
// its completion must neither clear the newer text nor
// repopulate the view with the older path.
supersede()
setLoading(false)
setPathDraft(event.target.value)
}}
{...compositionGuard}
onKeyDown={(event) => {
if (event.key === 'Enter' && !composingRef.current) {
event.preventDefault()
// Trim only detects a blank draft; the Host gets the
// original text — a real directory name may end in
// whitespace, and trimming would list its sibling.
if (pathDraft.trim() !== '') navigate(pathDraft)
}
if (event.key === 'Escape') {
event.stopPropagation()
// Cancel also withdraws a navigation the editor already
// launched: its late success must not jump to the
// cancelled path, so the pending request is superseded
// and the view leaves the loading state.
supersede()
setLoading(false)
setPathDraft(null)
setError(null)
// Editing may have superseded the selection's preview
// request; a selection with no preview would render a
// half-empty two-pane view, so cancel falls back to the
// single-pane level.
if (child === null) setSelected(null)
// With no level listed yet (the editor superseded the
// initial home listing), plain cancellation would leave a
// permanently blank picker: restart the home listing.
if (parent === null) navigate()
}
}}
setPathDraft(event.target.value)
}}
{...compositionGuard}
// Escape and focus-leave cancellation live on the card-scope
// wrapper above (they must work after focus Tabs onto the
// rows); this handler owns only submission.
onKeyDown={(event) => {
if (event.key === 'Enter' && !composingRef.current) {
event.preventDefault()
// Trim only detects a blank draft; the Host gets the
// original text — a real directory name may end in
// whitespace, and trimming would list its sibling.
if (pathDraft.trim() !== '') {
// Success will unmount the still-focused input; park
// focus on the returning crumb edit zone (a failure
// keeps the editor, so the flag waits until close).
refocusEditZone.current = true
navigate(pathDraft)
}
}
}}
/>
)}
</div>
</div>
<div className={css.content}>
<div className={css.millerRow} ref={millerRowRef}>
{parent !== null && (
<LevelColumn
entries={parent.entries}
selectedPath={selected?.path ?? null}
busy={parentInert}
onPick={select}
showHidden={showHidden}
filterPrefix={draftPrefixFor(parent, pathDraft)}
pathEditing={draftPending}
/>
)}
</div>
</div>
<div className={css.content}>
<div className={css.millerRow} ref={millerRowRef}>
{parent !== null && (
<LevelColumn
entries={parent.entries}
selectedPath={selected?.path ?? null}
busy={parentInert}
onPick={select}
wide={!twoPane}
/>
)}
{twoPane && <span className={css.divider} />}
{twoPane && child !== null && (
<LevelColumn
entries={child.entries}
selectedPath={null}
busy={parentInert}
onPick={advance}
wide={false}
/>
)}
</div>
{loading && <div className={css.status} role="status">{t('browser.loading')}</div>}
{/* The backend bounds a level at its complete-result limit; say so
{twoPane && <span className={css.divider} />}
{twoPane && child !== null && (
<LevelColumn
entries={child.entries}
selectedPath={null}
busy={parentInert}
onPick={advance}
showHidden={showHidden}
filterPrefix={draftPrefixFor(child, pathDraft)}
pathEditing={draftPending}
/>
)}
</div>
{loading && slowScan
&& <div className={clsx(css.status, css.loadingFloat)} role="status">{t('browser.loading')}</div>}
{/* The backend bounds a level at its complete-result limit; say so
* whenever a visible pane was cut instead of letting the tail of a
* huge directory go silently missing. */}
{(parent?.truncated === true || child?.truncated === true) && !loading
* huge directory go silently missing. The note describes the panes
* on screen, so an in-flight scan leaves it alone — hiding it while
* the stale view still shows the cut level would shift the columns
* on every navigation away from it. */}
{(parent?.truncated === true || child?.truncated === true)
&& <div className={css.status} role="status">{t('browser.truncated')}</div>}
{error !== null && <div className={css.error} role="alert">{error}</div>}
</div>
<div className={css.footerBar}>
<Button
variant="outline"
icon={<IconPlusOutline16 size={14} />}
disabled={parent === null || loading || parentInert || draftPending}
onClick={() => {
setFolderDraft('')
setCreateError(null)
}}
>
{t('browser.newFolder')}
</Button>
<span className={css.footerGap} />
<Button variant="outline" className={clsx(css.footerAction)} disabled={parentInert} onClick={onClose}>{t('browser.cancel')}</Button>
<Button
variant="primary"
className={clsx(css.footerAction)}
disabled={targetPath === null || loading || parentInert || draftPending}
/* v8 ignore next -- narrowing guard: Open disables while no target exists. */
onClick={() => { if (targetPath !== null) onOpen(targetPath) }}
>
{t('browser.open')}
</Button>
{error !== null && <div className={css.error} role="alert">{error}</div>}
</div>
<div className={css.footerBar}>
<Button
variant="outline"
icon={<IconPlusOutline16 size={14} />}
disabled={parent === null || loading || parentInert || draftPending}
onClick={() => {
setFolderDraft('')
setCreateError(null)
}}
>
{t('browser.newFolder')}
</Button>
<button
type="button"
className={clsx(css.showHiddenToggle, showHidden && css.showHiddenToggleActive)}
aria-pressed={showHidden}
disabled={parentInert}
// The toggle composes with the path editor (dot-led prefixes and
// this filter interleave): while editing, don't steal focus, so
// toggling never blur-cancels a draft mid-thought. Outside editing
// it keeps native focus behavior.
onMouseDown={draftPending ? (event) => { event.preventDefault() } : undefined}
onClick={() => { setShowHidden(prev => !prev) }}
>
{t('browser.showHidden')}
{/* Trailing check (Menu's selected vocabulary): the label never
* shifts when the pressed state toggles. */}
{showHidden && <IconCheckOutline16 size={14} />}
</button>
<span className={css.footerGap} />
<Button variant="outline" className={clsx(css.footerAction)} disabled={parentInert} onClick={onClose}>{t('browser.cancel')}</Button>
<Button
variant="primary"
className={clsx(css.footerAction)}
disabled={targetPath === null || loading || parentInert || draftPending}
/* v8 ignore next -- narrowing guard: Open disables while no target exists. */
onClick={() => { if (targetPath !== null) onOpen(targetPath) }}
>
{t('browser.open')}
</Button>
</div>
</div>
{/* Nested create dialog (figma 813:23278): names one folder inside the target. */}
<Modal

View File

@@ -46,6 +46,7 @@ export function apply(ctx: ClientContext): void {
'browser.editPath': '编辑路径',
'browser.loading': '加载中…',
'browser.truncated': '文件夹过多,仅显示开头部分。',
'browser.showHidden': '显示隐藏文件',
}],
['en', {
'browser.title': 'Select Workspace Directory',
@@ -60,6 +61,7 @@ export function apply(ctx: ClientContext): void {
'browser.editPath': 'Edit path',
'browser.loading': 'Loading…',
'browser.truncated': 'Too many folders to list; only the beginning is shown.',
'browser.showHidden': 'Show hidden files',
}],
]
try {

View File

@@ -162,6 +162,7 @@ describe('directory-picker-browse client half', () => {
// zh is the shipped default locale.
expect(injected.t('browser.title')).toBe('选择工作区目录')
expect(injected.t('browser.newFolder')).toBe('新建文件夹')
expect(injected.t('browser.showHidden')).toBe('显示隐藏文件')
})
it('drives the injected browse calls through the hole entry', async () => {

View File

@@ -29,6 +29,25 @@ function listingFor(path?: string): DirectoryListing {
],
truncated: false,
},
'/': {
path: '/',
home: HOME,
crumbs: [{ name: '/', path: '/', hidden: false }],
entries: [{ name: 'home', path: '/home', hidden: false }],
truncated: false,
},
[`${HOME}/.config`]: {
path: `${HOME}/.config`,
home: HOME,
crumbs: [
{ name: '/', path: '/', hidden: false },
{ name: 'home', path: '/home', hidden: false },
{ name: 'u', path: HOME, hidden: false },
{ name: '.config', path: `${HOME}/.config`, hidden: true },
],
entries: [],
truncated: false,
},
[DOCS]: {
path: DOCS,
home: HOME,
@@ -92,6 +111,12 @@ function rowButton(item: HTMLElement): HTMLButtonElement {
}
describe('DirectoryBrowser', () => {
it('renders nothing and launches no listing while initially closed', () => {
const b = mount({ open: false })
expect(screen.queryByRole('dialog')).toBeNull()
expect(b.listDirectory).not.toHaveBeenCalled()
})
it('opens at the Host home as one wide column, hides hidden entries, and roots the crumbs at Home', async () => {
const b = mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
@@ -103,6 +128,28 @@ describe('DirectoryBrowser', () => {
expect(screen.queryByRole('button', { name: '/' })).toBeNull()
})
it('shows hidden entries when the toggle is on and hides them again on close', async () => {
const b = mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
expect(screen.queryByText('.config')).toBeNull()
// The fixed-label toggle reports its state through aria-pressed. Its
// mousedown never steals focus (so it composes with the path editor).
const toggle = screen.getByRole('button', { name: 'browser.showHidden' })
expect(toggle.getAttribute('aria-pressed')).toBe('false')
fireEvent.mouseDown(toggle)
fireEvent.click(toggle)
expect(toggle.getAttribute('aria-pressed')).toBe('true')
expect(screen.getByText('.config')).toBeTruthy()
fireEvent.click(toggle)
expect(toggle.getAttribute('aria-pressed')).toBe('false')
expect(screen.queryByText('.config')).toBeNull()
// Close resets the toggle.
b.view.rerender(<DirectoryBrowser {...b.props} open={false} />)
b.view.rerender(<DirectoryBrowser {...b.props} open />)
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
expect(screen.queryByText('.config')).toBeNull()
})
it('selects a row into the two-pane view: children preview right, crumbs follow the selection', async () => {
const b = mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
@@ -153,7 +200,7 @@ describe('DirectoryBrowser', () => {
expect(signals[2]?.aborted).toBe(true)
})
it('jumps back through a crumb into a fresh single-column level', async () => {
it('a crumb jump to the display root (home) lands the single wide level', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(rowButton(screen.getByRole('listitem')))
@@ -164,6 +211,393 @@ describe('DirectoryBrowser', () => {
expect(rowButton(screen.getByRole('listitem')).getAttribute('aria-current')).toBeNull()
})
it('a crumb jump away from the root lands two-pane with the target selected', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(rowButton(screen.getByRole('listitem')))
await waitFor(() => { expect(columns()).toHaveLength(2) })
fireEvent.click(rowButton(within(columns()[1]!).getByRole('listitem')))
await waitFor(() => { expect(screen.getByRole('button', { name: 'harness' })).toBeTruthy() })
// Jumping to the Documents crumb is a step BACK one pane, not a
// collapse: Documents stays selected in the home level, its children
// stay on the right.
fireEvent.click(screen.getByRole('button', { name: 'Documents' }))
await waitFor(() => {
expect(rowButton(within(columns()[0]!).getByRole('listitem')).getAttribute('aria-current')).toBe('true')
})
expect(columns()).toHaveLength(2)
expect(within(columns()[0]!).getByText('Documents')).toBeTruthy()
expect(within(columns()[1]!).getByText('harness')).toBeTruthy()
})
it('a navigation to the filesystem root keeps the single wide level', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: '/' } })
fireEvent.keyDown(screen.getByLabelText('browser.editPath'), { key: 'Enter' })
// A one-crumb chain has no parent level to show on the left.
await waitFor(() => { expect(screen.getByRole('listitem').textContent).toBe('home') })
expect(columns()).toHaveLength(1)
})
it('lands the target single-pane at the wait bound, aborts a superseded parent leg on the wire, and drops its late resolution', async () => {
vi.useFakeTimers()
try {
const signals: (AbortSignal | undefined)[] = []
const settlers: ((value: DirectoryListing) => void)[] = []
// Only the FIRST explicit HOME request (the parent leg) hangs; the
// later home crumb jump lists normally.
let homeCalls = 0
const listDirectory = vi.fn((path?: string, signal?: AbortSignal) => {
signals.push(signal)
if (path === HOME && ++homeCalls === 1) {
return new Promise<DirectoryListing>((resolve) => { settlers.push(resolve) })
}
return Promise.resolve(listingFor(path))
})
mount({ listDirectory })
await act(async () => {})
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: DOCS } })
fireEvent.keyDown(screen.getByLabelText('browser.editPath'), { key: 'Enter' })
// The target settled but the parent leg hangs: inside the wait bound
// nothing commits yet.
await act(async () => {})
expect(settlers).toHaveLength(1)
expect(screen.getByLabelText('browser.editPath', { selector: 'input' })).toBeTruthy()
// The wait bound expires: the target commits alone — editor closed,
// single-pane DOCS level.
await act(async () => { vi.advanceTimersByTime(200) })
expect(screen.getByRole('listitem').textContent).toBe('harness')
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
expect(columns()).toHaveLength(1)
// A newer jump aborts the pending parent leg ON THE WIRE, not merely
// dropping its settlement.
fireEvent.click(screen.getByRole('button', { name: 'browser.home' }))
expect(signals[2]?.aborted).toBe(true)
await act(async () => {})
expect(screen.getByRole('listitem').textContent).toBe('Documents')
// Its late resolution changes nothing either.
await act(async () => { settlers[0]!(listingFor(HOME)) })
expect(columns()).toHaveLength(1)
expect(rowButton(screen.getByRole('listitem')).getAttribute('aria-current')).toBeNull()
} finally {
vi.useRealTimers()
}
})
/**
* Listing fake whose explicit-path scans stay pending until the test
* settles them by path; the absent-path form (the initial home listing)
* resolves normally so mounting is a one-flush setup.
*/
function manualLister() {
const settlers = new Map<string, (value: DirectoryListing) => void>()
const listDirectory = vi.fn((path?: string, _signal?: AbortSignal) => {
if (path === undefined) return Promise.resolve(listingFor(path))
return new Promise<DirectoryListing>((resolve) => { settlers.set(path, resolve) })
})
return { settlers, listDirectory }
}
it('lands a navigation as ONE two-pane frame: the stale view holds until both legs arrive', async () => {
vi.useFakeTimers()
try {
const { settlers, listDirectory } = manualLister()
mount({ listDirectory })
await act(async () => {})
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: DOCS } })
fireEvent.keyDown(screen.getByLabelText('browser.editPath'), { key: 'Enter' })
// The target settles while the parent leg is still in flight: nothing
// commits yet — the editor stays open over the stale home level, and no
// single-pane DOCS frame ever renders.
await act(async () => { settlers.get(DOCS)!(listingFor(DOCS)) })
expect(screen.getByLabelText('browser.editPath', { selector: 'input' })).toBeTruthy()
expect(screen.queryByText('harness')).toBeNull()
// The parent leg settles inside the wait bound: one commit straight to
// the two-pane landing, editor closed.
await act(async () => { settlers.get(HOME)!(listingFor(HOME)) })
expect(columns()).toHaveLength(2)
expect(rowButton(within(columns()[0]!).getByRole('listitem')).getAttribute('aria-current')).toBe('true')
expect(within(columns()[0]!).getByText('Documents')).toBeTruthy()
expect(within(columns()[1]!).getByText('harness')).toBeTruthy()
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
// The wait-bound timer firing after the landing is a no-op.
await act(async () => { vi.advanceTimersByTime(200) })
expect(columns()).toHaveLength(2)
} finally {
vi.useRealTimers()
}
})
it('a stalled parent leg lands the target alone at the wait bound, then upgrades in place', async () => {
vi.useFakeTimers()
try {
const { settlers, listDirectory } = manualLister()
mount({ listDirectory })
await act(async () => {})
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: DOCS } })
fireEvent.keyDown(screen.getByLabelText('browser.editPath'), { key: 'Enter' })
// The target can consume most of the outer scan's silence window.
await act(async () => { vi.advanceTimersByTime(250) })
await act(async () => { settlers.get(DOCS)!(listingFor(DOCS)) })
// Its parent leg gets a fresh silence window. Crossing the original
// scan's 300ms deadline therefore cannot flash the indicator during the
// bounded landing wait.
await act(async () => { vi.advanceTimersByTime(199) })
expect(screen.queryByText('browser.loading')).toBeNull()
// The parent leg outlives PARENT_LEG_WAIT_MS: the target lands alone.
await act(async () => { vi.advanceTimersByTime(1) })
expect(columns()).toHaveLength(1)
expect(screen.getByRole('listitem').textContent).toBe('harness')
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
// The late parent leg still upgrades the landing in place, exactly as
// if it had made the bound. (Reopening the editor meanwhile would
// supersede the upgrade — the editor-open handler withdraws pending
// listings — so a late upgrade can never close a resumed draft.)
await act(async () => { settlers.get(HOME)!(listingFor(HOME)) })
expect(columns()).toHaveLength(2)
expect(rowButton(within(columns()[0]!).getByRole('listitem')).getAttribute('aria-current')).toBe('true')
expect(within(columns()[1]!).getByText('harness')).toBeTruthy()
} finally {
vi.useRealTimers()
}
})
it('Escape inside the landing window withdraws the submitted navigation', async () => {
vi.useFakeTimers()
try {
const { settlers, listDirectory } = manualLister()
mount({ listDirectory })
await act(async () => {})
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
fireEvent.change(input, { target: { value: DOCS } })
fireEvent.keyDown(input, { key: 'Enter' })
await act(async () => { settlers.get(DOCS)!(listingFor(DOCS)) })
// Nothing has committed yet; Escape supersedes the landing entirely.
fireEvent.keyDown(input, { key: 'Escape' })
await act(async () => { vi.advanceTimersByTime(200) })
expect(columns()).toHaveLength(1)
expect(screen.queryByText('harness')).toBeNull()
expect(within(columns()[0]!).getByText('Documents')).toBeTruthy()
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
} finally {
vi.useRealTimers()
}
})
it('shows the loading indicator only once a scan outlives its silence window, floating over the stale view', async () => {
vi.useFakeTimers()
try {
// The home level is truncated so its note is on screen when the slow
// scan starts: dropping the note's old !loading guard means it must
// keep rendering through the scan, coexisting with the indicator.
const settlers = new Map<string, (value: DirectoryListing) => void>()
const listDirectory = vi.fn((path?: string, _signal?: AbortSignal) => {
if (path === undefined) return Promise.resolve({ ...listingFor(path), truncated: true })
return new Promise<DirectoryListing>((resolve) => { settlers.set(path, resolve) })
})
mount({ listDirectory })
await act(async () => {})
expect(screen.queryByText('browser.loading')).toBeNull()
expect(screen.getByText('browser.truncated')).toBeTruthy()
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: DOCS } })
fireEvent.keyDown(screen.getByLabelText('browser.editPath'), { key: 'Enter' })
// In flight but still inside the silence window: no indicator, and the
// stale level's truncated note stays put (no layout churn on launch).
expect(screen.queryByText('browser.loading')).toBeNull()
expect(screen.getByText('browser.truncated')).toBeTruthy()
await act(async () => { vi.advanceTimersByTime(300) })
// Past it: the indicator floats while the stale level — truncated note
// included — keeps rendering beneath it.
expect(screen.getByText('browser.loading')).toBeTruthy()
expect(screen.getByText('browser.truncated')).toBeTruthy()
expect(screen.getByText('Documents')).toBeTruthy()
// Landing (both legs) retires the indicator with the scan, and the
// fresh listings' own truncated state replaces the stale note.
await act(async () => { settlers.get(DOCS)!(listingFor(DOCS)) })
await act(async () => { settlers.get(HOME)!(listingFor(HOME)) })
expect(screen.queryByText('browser.loading')).toBeNull()
expect(screen.queryByText('browser.truncated')).toBeNull()
expect(columns()).toHaveLength(2)
} finally {
vi.useRealTimers()
}
})
it('restarts the silence window when a row pick supersedes a pending scan', async () => {
vi.useFakeTimers()
try {
const pending: ((value: DirectoryListing) => void)[] = []
const listDirectory = vi.fn((path?: string, _signal?: AbortSignal) => {
if (path === undefined) return Promise.resolve(listingFor(path))
return new Promise<DirectoryListing>((resolve) => { pending.push(resolve) })
})
mount({ listDirectory })
await act(async () => {})
const documents = rowButton(screen.getByRole('listitem'))
fireEvent.click(documents)
await act(async () => { vi.advanceTimersByTime(300) })
expect(screen.getByText('browser.loading')).toBeTruthy()
// The same row remains actionable while its preview is pending. A second
// pick starts a new listing without a false `loading` edge.
fireEvent.click(documents)
expect(screen.queryByText('browser.loading')).toBeNull()
await act(async () => { vi.advanceTimersByTime(299) })
expect(screen.queryByText('browser.loading')).toBeNull()
await act(async () => { vi.advanceTimersByTime(1) })
expect(screen.getByText('browser.loading')).toBeTruthy()
await act(async () => { pending.at(-1)!(listingFor(DOCS)) })
} finally {
vi.useRealTimers()
}
})
it('a close mid-scan resets the slow-scan gate: reopening waits a fresh silence window', async () => {
vi.useFakeTimers()
try {
// Every home listing hangs: the initial open's scan is the one the
// close interrupts, and the reopen's scan proves the fresh window.
const settlers: ((value: DirectoryListing) => void)[] = []
const listDirectory = vi.fn((_path?: string, _signal?: AbortSignal) =>
new Promise<DirectoryListing>((resolve) => { settlers.push(resolve) }))
const { view, props } = mount({ listDirectory })
await act(async () => { vi.advanceTimersByTime(300) })
expect(screen.getByText('browser.loading')).toBeTruthy()
// Close while the scan is in flight, then reopen: the first frame must
// wait out a fresh silence window, not inherit the armed indicator.
view.rerender(<DirectoryBrowser {...props} open={false} />)
view.rerender(<DirectoryBrowser {...props} open />)
await act(async () => {})
expect(screen.queryByText('browser.loading')).toBeNull()
await act(async () => { vi.advanceTimersByTime(300) })
expect(screen.getByText('browser.loading')).toBeTruthy()
// The reopened scan settles normally.
await act(async () => { settlers.at(-1)!(listingFor(undefined)) })
expect(screen.queryByText('browser.loading')).toBeNull()
expect(screen.getByText('Documents')).toBeTruthy()
} finally {
vi.useRealTimers()
}
})
it('keeps the single-pane landing when the truncated parent level lacks the target', async () => {
const listDirectory = vi.fn(async (path?: string) => {
// The parent leg names HOME explicitly; serve it a truncated window
// that misses Documents (the initial open uses the absent-path form).
if (path === HOME) return { ...listingFor(HOME), entries: [], truncated: true }
return listingFor(path)
})
mount({ listDirectory })
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: DOCS } })
fireEvent.keyDown(screen.getByLabelText('browser.editPath'), { key: 'Enter' })
await waitFor(() => { expect(screen.getByRole('listitem').textContent).toBe('harness') })
// The upgrade would orphan the selection (no source row): it stays off.
await act(async () => {})
expect(columns()).toHaveLength(1)
expect(screen.queryByText('browser.truncated')).toBeNull()
})
it('anchors the upgrade on the parent level actual entry under Windows case folding', async () => {
const ROOT = 'C:\\'
const TYPED = 'c:\\users'
const winRoot: DirectoryListing = {
path: ROOT,
home: ROOT,
crumbs: [{ name: 'C:\\', path: ROOT, hidden: false }],
entries: [{ name: 'Users', path: 'C:\\Users', hidden: false }],
truncated: false,
}
const winUsers: DirectoryListing = {
path: TYPED,
home: ROOT,
crumbs: [{ name: 'C:\\', path: ROOT, hidden: false }, { name: 'users', path: TYPED, hidden: false }],
entries: [],
truncated: false,
}
mount({ listDirectory: vi.fn(async (path?: string) => (path === TYPED ? winUsers : winRoot)) })
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: TYPED } })
fireEvent.keyDown(screen.getByLabelText('browser.editPath'), { key: 'Enter' })
// The typed case differs from the real entry; the upgrade selects the
// parent level's ACTUAL entry so aria-current and exemptions hold.
await waitFor(() => {
expect(rowButton(within(columns()[0]!).getByRole('listitem')).getAttribute('aria-current')).toBe('true')
})
expect(within(columns()[0]!).getByText('Users')).toBeTruthy()
})
it('re-parks focus on the edit zone when a failed pick unmounts a dot-revealed row', async () => {
const listDirectory = vi.fn(async (path?: string) => {
if (path === `${HOME}/.config`) {
throw new DirectoryBrowseError({ code: 'directory-unreadable', message: 'denied', details: { path } })
}
return listingFor(path)
})
mount({ listDirectory })
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: `${HOME}/.co` } })
const row = rowButton(screen.getByRole('listitem'))
fireEvent.mouseDown(row)
fireEvent.click(row)
// The failed selection re-hides the picked row; focus fell to body and
// re-parks on the crumb edit zone.
await screen.findByRole('alert')
expect(screen.queryByText('.config')).toBeNull()
expect(document.activeElement).toBe(screen.getByRole('button', { name: 'browser.editPath' }))
})
it('leaves focus on a surviving row when its pick fails', async () => {
const listDirectory = vi.fn(async (path?: string) => {
if (path === DOCS) {
throw new DirectoryBrowseError({ code: 'directory-unreadable', message: 'denied', details: { path } })
}
return listingFor(path)
})
mount({ listDirectory })
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: `${HOME}/do` } })
const row = rowButton(screen.getByRole('listitem'))
row.focus()
fireEvent.mouseDown(row)
fireEvent.click(row)
// Documents survives the cleared selection (it is not hidden): the
// user's focus on it is not yanked to the edit zone.
await screen.findByRole('alert')
expect(document.activeElement).toBe(row)
})
it('falls back to the single-pane landing when the parent leg of a navigation fails', async () => {
const listDirectory = vi.fn(async (path?: string) => {
// The initial open lists home through the absent-path form; only the
// parent leg names HOME explicitly.
if (path === HOME) {
throw new DirectoryBrowseError({ code: 'directory-unreadable', message: 'parent gone', details: { path } })
}
return listingFor(path)
})
mount({ listDirectory })
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.change(screen.getByLabelText<HTMLInputElement>('browser.editPath'), { target: { value: DOCS } })
fireEvent.keyDown(screen.getByLabelText('browser.editPath'), { key: 'Enter' })
// The target listed fine; the failed parent leg neither blocks the
// landing nor surfaces an error for a level nobody asked to see.
await waitFor(() => { expect(screen.getByRole('listitem').textContent).toBe('harness') })
expect(columns()).toHaveLength(1)
expect(screen.queryByRole('alert')).toBeNull()
})
it('opens the selection, else the listed level; Cancel closes; busy freezes Open', async () => {
const b = mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
@@ -186,18 +620,221 @@ describe('DirectoryBrowser', () => {
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
expect(input.value).toBe(HOME)
// The editor seeds with a trailing separator so typing continues into
// child names.
expect(input.value).toBe(`${HOME}/`)
fireEvent.change(input, { target: { value: DOCS } })
fireEvent.keyDown(input, { key: 'Enter' })
await waitFor(() => { expect(screen.getByRole('listitem').textContent).toBe('harness') })
expect(columns()).toHaveLength(1)
// Away from the root a navigation lands two-pane: the target selected
// in its parent level, its own children on the right.
await waitFor(() => { expect(columns()).toHaveLength(2) })
expect(rowButton(within(columns()[0]!).getByRole('listitem')).getAttribute('aria-current')).toBe('true')
expect(within(columns()[1]!).getByText('harness')).toBeTruthy()
// The submitted navigation unmounted the focused input; focus parks on
// the crumb edit zone that replaced it.
expect(document.activeElement).toBe(screen.getByRole('button', { name: 'browser.editPath' }))
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const again = screen.getByLabelText<HTMLInputElement>('browser.editPath')
fireEvent.change(again, { target: { value: ' ' } })
fireEvent.keyDown(again, { key: 'Enter' })
expect(b.listDirectory).toHaveBeenCalledTimes(2)
// Initial home + the DOCS target leg + its parent leg; the blank draft
// added none.
expect(b.listDirectory).toHaveBeenCalledTimes(3)
fireEvent.keyDown(again, { key: 'Escape' })
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
// Escape with focus in the input parks focus on the returning edit zone.
expect(document.activeElement).toBe(screen.getByRole('button', { name: 'browser.editPath' }))
})
it('prefix-filters the listed level from the draft tail, dot revealing hidden matches', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
// The seeded empty segment leaves the level as-is: hidden stays hidden.
expect(screen.getByRole('listitem').textContent).toBe('Documents')
// Case-insensitive prefix narrows the rows.
fireEvent.change(input, { target: { value: `${HOME}/do` } })
expect(screen.getByRole('listitem').textContent).toBe('Documents')
// A dot-led prefix names hidden entries, so it reveals the match.
fireEvent.change(input, { target: { value: `${HOME}/.co` } })
expect(screen.getByRole('listitem').textContent).toBe('.config')
// A prefix matching nothing empties the level (no stale rows linger).
fireEvent.change(input, { target: { value: `${HOME}/zzz` } })
expect(screen.queryByRole('listitem')).toBeNull()
// A draft naming some other directory (or none) leaves the level whole.
fireEvent.change(input, { target: { value: 'no-separator' } })
expect(screen.getByRole('listitem').textContent).toBe('Documents')
})
it('filters the child pane in two-pane mode and follows the draft back up a level', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(rowButton(screen.getByRole('listitem')))
await waitFor(() => { expect(columns()).toHaveLength(2) })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
// The seed comes from the selection, so the draft tail addresses the
// RIGHT pane (the selection's children).
expect(input.value).toBe(`${DOCS}/`)
fireEvent.change(input, { target: { value: `${DOCS}/h` } })
expect(within(columns()[1]!).getByText('harness')).toBeTruthy()
fireEvent.change(input, { target: { value: `${DOCS}/zzz` } })
expect(within(columns()[1]!).queryAllByRole('listitem')).toHaveLength(0)
expect(within(columns()[0]!).getByText('Documents')).toBeTruthy()
// Erasing back into the parent's own path moves the filter to the LEFT
// pane and releases the right one. The selected row is exempt (it
// anchors the two-pane view), so it alone survives the miss.
fireEvent.change(input, { target: { value: `${HOME}/zz` } })
expect(within(columns()[0]!).getAllByRole('listitem').map(item => item.textContent)).toEqual(['Documents'])
expect(within(columns()[1]!).getByText('harness')).toBeTruthy()
})
it('keeps the draft and filter through window focus loss and in-dialog focus moves', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
fireEvent.change(input, { target: { value: `${HOME}/do` } })
// A blur while the document itself lost focus (window switch, dev-tools
// focus) must not discard the draft: value and filter both survive.
const hasFocus = vi.spyOn(document, 'hasFocus').mockReturnValue(false)
fireEvent.focusOut(input)
hasFocus.mockRestore()
expect(screen.getByLabelText<HTMLInputElement>('browser.editPath', { selector: 'input' }).value).toBe(`${HOME}/do`)
expect(screen.getByRole('listitem').textContent).toBe('Documents')
// A keyboard focus move that stays inside the dialog (Tab onto the
// filtered row) keeps the draft too — the results stay reachable.
fireEvent.focusOut(input, { relatedTarget: rowButton(screen.getByRole('listitem')) })
expect(screen.getByLabelText<HTMLInputElement>('browser.editPath', { selector: 'input' }).value).toBe(`${HOME}/do`)
// Toggling show-hidden mid-edit suppresses focus steal: the draft and
// its filter survive the toggle in both directions.
const toggle = screen.getByRole('button', { name: 'browser.showHidden' })
fireEvent.mouseDown(toggle)
fireEvent.click(toggle)
expect(toggle.getAttribute('aria-pressed')).toBe('true')
expect(screen.getByLabelText<HTMLInputElement>('browser.editPath', { selector: 'input' }).value).toBe(`${HOME}/do`)
expect(screen.getByRole('listitem').textContent).toBe('Documents')
// Focus landing outside the dialog cancels like Escape — even when the
// departure happens from a row the user had Tabbed onto, not the input
// (the observer lives on the card scope, not the input).
fireEvent.focusOut(rowButton(screen.getByRole('listitem')), { relatedTarget: document.body })
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
// Outside editing the card-scope observer is inert.
fireEvent.focusOut(screen.getByRole('button', { name: 'browser.showHidden' }))
expect(screen.getByRole('button', { name: 'browser.editPath' })).toBeTruthy()
})
it('Escape with focus on a filtered row collapses the editor, not the dialog', async () => {
const b = mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
fireEvent.change(input, { target: { value: `${HOME}/do` } })
// Tab parked focus on the result row; Escape must still mean "leave
// path editing", not "close the whole dialog".
const row = rowButton(screen.getByRole('listitem'))
row.focus()
fireEvent.keyDown(row, { key: 'Escape' })
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
expect(b.onClose).not.toHaveBeenCalled()
// Focus was already on a surviving row, so nothing re-parks it.
expect(document.activeElement).toBe(row)
// With no draft left, Escape falls through to the Modal and closes.
fireEvent.keyDown(row, { key: 'Escape' })
expect(b.onClose).toHaveBeenCalledTimes(1)
})
it('a picked dot-revealed hidden row stays visible as the selection', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
fireEvent.change(input, { target: { value: `${HOME}/.co` } })
const row = rowButton(screen.getByRole('listitem'))
expect(row.textContent).toBe('.config')
fireEvent.mouseDown(row)
fireEvent.click(row)
// The pick cleared the draft (and with it the dot-reveal), but the
// selection is exempt from the hidden filter: the anchor row survives.
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
await waitFor(() => { expect(columns()).toHaveLength(2) })
expect(within(columns()[0]!).getByText('.config')).toBeTruthy()
})
it('picking a filtered row adopts it and closes the path editor', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
fireEvent.change(input, { target: { value: `${HOME}/do` } })
// The row suppresses focus steal on mousedown (no blur-cancel unmounts
// the filtered rows mid-gesture), then the click both selects the row
// and closes the editor.
const row = rowButton(screen.getByRole('listitem'))
fireEvent.mouseDown(row)
fireEvent.click(row)
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
// Focus parks on the picked row (the editor's input just unmounted and
// the Modal has no focus trap to catch a fall to body).
expect(document.activeElement).toBe(row)
await waitFor(() => { expect(columns()).toHaveLength(2) })
expect(screen.getByRole('button', { name: 'browser.home' })).toBeTruthy()
})
it('a right-pane pick while editing parks focus on the advanced selection', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(rowButton(screen.getByRole('listitem')))
await waitFor(() => { expect(columns()).toHaveLength(2) })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
fireEvent.change(input, { target: { value: `${DOCS}/h` } })
// The advance replaces BOTH panes (the picked button's own column
// unmounts), so focus is re-parked on the selection's aria-current row
// in the freshly rendered left pane rather than the clicked node.
const row = rowButton(within(columns()[1]!).getByRole('listitem'))
fireEvent.mouseDown(row)
fireEvent.click(row)
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
await waitFor(() => { expect(document.activeElement?.textContent).toBe('harness') })
expect(document.activeElement?.getAttribute('aria-current')).toBe('true')
})
it('seeds and filters with backslashes on a Windows-rooted listing', async () => {
const ROOT = 'C:\\'
const windowsListing: DirectoryListing = {
path: ROOT,
home: ROOT,
crumbs: [{ name: 'C:\\', path: ROOT, hidden: false }],
entries: [
{ name: 'Program Files', path: `${ROOT}Program Files`, hidden: false },
{ name: 'Users', path: `${ROOT}Users`, hidden: false },
],
truncated: false,
}
mount({ listDirectory: vi.fn(async () => windowsListing) })
await waitFor(() => { expect(screen.getAllByRole('listitem')).toHaveLength(2) })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
// The root already ends in its separator: no doubled backslash.
expect(input.value).toBe(ROOT)
fireEvent.change(input, { target: { value: `${ROOT}u` } })
expect(screen.getByRole('listitem').textContent).toBe('Users')
})
it('clicking away from the path editor cancels it back to the crumb view', async () => {
mount()
await waitFor(() => { expect(screen.getByRole('listitem')).toBeTruthy() })
fireEvent.click(screen.getByRole('button', { name: 'browser.editPath' }))
const input = screen.getByLabelText<HTMLInputElement>('browser.editPath')
fireEvent.change(input, { target: { value: '/somewhere/else' } })
// Focus moving anywhere outside the editor abandons the draft like Escape.
fireEvent.focusOut(input)
expect(screen.queryByLabelText('browser.editPath', { selector: 'input' })).toBeNull()
// The crumb view is back and the abandoned draft was never navigated to.
expect(screen.getByRole('button', { name: 'browser.editPath' })).toBeTruthy()
expect(screen.getByRole('listitem').textContent).toBe('Documents')
})
it('restarts the home listing when Escape cancels an edit opened before any level listed', async () => {
@@ -713,11 +1350,13 @@ describe('DirectoryBrowser', () => {
b.listDirectory.mockReturnValueOnce(slow)
fireEvent.click(screen.getByRole('button', { name: 'browser.home' }))
fireEvent.click(within(screen.getByRole('navigation')).getByRole('button', { name: 'Documents' }))
await waitFor(() => { expect(screen.getByRole('listitem').textContent).toBe('harness') })
// The newer jump lands two-pane: Documents selected at home, children right.
await waitFor(() => { expect(within(columns()[1]!).getByText('harness')).toBeTruthy() })
resolveSlow(listingFor(undefined))
await new Promise(settle => setTimeout(settle, 0))
// The stale home listing did not replace the newer Documents level.
expect(screen.getByRole('listitem').textContent).toBe('harness')
// The stale home listing did not replace the newer Documents landing.
expect(columns()).toHaveLength(2)
expect(within(columns()[1]!).getByText('harness')).toBeTruthy()
})
it('names the create target by its path when the level reports no crumbs', async () => {

View File

@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write packages/pty/README.md
README.md: a9121455519a5f83a63a005cb857fec0f0e06b92
README.zh.md: 4e36ef934d31861d72748b9b9585695f86749171
README.md: e54dcf64db083665f37b7dc19a7a92e21494442b
README.zh.md: 0bb7e565a5799002ffb8b74dce12e129a7b2665e

View File

@@ -9,5 +9,6 @@ English | [中文](README.zh.md)
| [`pty`](pty/README.md) (`@deepseek-ai/dsh-pty`) | Backend registry, branded ids, exact-Agent ownership, session operations, and awaited cleanup | `ctx.pty` |
| `pty-local` (`@deepseek-ai/dsh-pty-local`) | Local `node-pty` backend, readiness detection, bounded terminal state, sandboxing, and process-session supervision | registers on `ctx.pty` |
| `tool-pty` (`@deepseek-ai/dsh-tool-pty`) | Six model-facing tools and generic task integration for background sends | registers on `ctx.tools` |
| `tool-bash-persistent` (`@deepseek-ai/dsh-tool-bash-persistent`) | One model-facing `bash` backed by an owner-scoped reusable PTY shell | consumes `ctx.pty`, registers on `ctx.tools` |
The design and deferred boundaries live in the [persistent PTY Agent Note](../../.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.md).

View File

@@ -9,5 +9,6 @@
| [`pty`](pty/README.md)`@deepseek-ai/dsh-pty` | 后端注册表、品牌化 id、精确到 agent智能体的所有权、会话操作与等待清理完成的机制 | `ctx.pty` |
| `pty-local``@deepseek-ai/dsh-pty-local` | 本地 `node-pty` 后端、就绪检测、有界终端状态、沙箱与进程会话监管 | 注册到 `ctx.pty` |
| `tool-pty``@deepseek-ai/dsh-tool-pty` | 6 个面向模型的工具,以及用于后台发送的通用任务集成 | 注册到 `ctx.tools` |
| `tool-bash-persistent``@deepseek-ai/dsh-tool-bash-persistent` | 一个由所有者隔离可复用 PTY shell 支撑的模型可见 `bash` | 消费 `ctx.pty`,注册到 `ctx.tools` |
设计与暂缓边界记录在[持久 PTY Agent Noteagent 决策记录)](../../.agents/notes/implemented/feature/2026-07-16-persistent-pty-sessions.md) 中。

View File

@@ -0,0 +1,6 @@
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write packages/pty/tool-bash-persistent/README.md
README.md: cfb3acf41803f5b06ecb3f56ee8ffc29ce9147b1
README.zh.md: a525485933344a8f9bc218456c099efdef572902

View File

@@ -0,0 +1,50 @@
# @deepseek-ai/dsh-tool-bash-persistent
English | [中文](README.zh.md)
Model-facing `bash(command)` backed by one owner-scoped `ctx.pty` shell. The package owns the tool contract and shell reuse; deployments select the PTY backend and sandbox policy.
## Config
| Key | Default | Meaning |
|---|---:|---|
| `backendType` | `shell` | Registered PTY backend used for each Agent shell. |
| `timeoutMs` | `300000` | Wall-clock limit for one command; timeout closes the shell. |
| `maxOutputChars` | `16000` | Maximum retained command-output characters; fixed diagnostics are added afterward. |
| `description` | Persistent-shell description | Model-facing environment contract. |
## Model Experience
### Tool schema
#### What the model sees
The generated [`bash` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-bash-persistent), including the configured `description`. The plugin contributes no standalone system-prompt section; the deployment owns persona and environment guidance.
#### Token effect
Fixed schema cost while `bash` is visible.
#### KV Cache effect
Prefix-stable while the configured description and schema remain unchanged.
### Tool results
#### What the model sees
Commands share one shell per Agent, so cwd, exported variables, activated environments, functions, and background jobs persist across calls. Results exclude private completion markers and the shell prompt. A nonzero wrapped command appends `[exit code: N]`; a shell that exits before reporting that status instead appends `[shell exited: code N]`, `[shell killed by signal: SIG]`, or `[shell exited]` when the backend supplies neither, then resets and tells the model that the next call starts fresh. Long output keeps the earliest retained prefix plus a clipping notice. If the PTY has already dropped that prefix, the result says so explicitly instead of presenting a tail as complete output. Timeout returns bounded partial output, closes the uncertain shell, and reports the reset.
#### Token effect
Data-dependent. `maxOutputChars` bounds retained command output; fixed clipping, lost-prefix, status, timeout, and reset diagnostics can extend the result.
#### KV Cache effect
Append-only tool results follow the reusable request prefix.
## Known Limitations and Deferred Work
- The tool requires an owning Agent and a real PTY backend.
- Explicit `exit` and timeout discard shell state. Cancellation also resets and discards the result, even when a complete status marker is already observable; the next call starts a fresh shell.
- Environment facts such as network access and package mirrors belong in the configured `description`, not this package's default.

View File

@@ -0,0 +1,50 @@
# @deepseek-ai/dsh-tool-bash-persistent
[English](README.md) | 中文
模型可见的 `bash(command)`,底层复用一个按所有者隔离的 `ctx.pty` shell。该包拥有工具契约和 shell 复用PTY 后端与沙箱策略由部署选择。
## 配置
| 键 | 默认值 | 含义 |
|---|---:|---|
| `backendType` | `shell` | 每个 Agent shell 使用的已注册 PTY 后端。 |
| `timeoutMs` | `300000` | 单条命令的墙钟时间上限;超时会关闭 shell。 |
| `maxOutputChars` | `16000` | 命令输出最多保留的字符数;固定诊断会在此后追加。 |
| `description` | 持久 shell 描述 | 面向模型的环境契约。 |
## 模型体验
### 工具 schema
#### 模型所见
生成的 [`bash` schema](../../../docs/tool-catalog.md#deepseek-aidsh-tool-bash-persistent),其中包含配置的 `description`。本插件不贡献独立系统提示词段persona 与环境指导由部署负责。
#### Token 影响
`bash` 可见时产生固定的 schema 成本。
#### KV Cache 影响
配置的描述与 schema 不变时前缀稳定。
### 工具结果
#### 模型所见
每个 Agent 的命令共享一个 shell因此 cwd、导出的环境变量、已激活环境、函数和后台任务会跨调用保留。结果不包含私有完成标记和 shell 提示符。经封装的命令以非零状态结束时,结果会追加 `[exit code: N]`;若 shell 在报告该状态前退出,则改为追加 `[shell exited: code N]``[shell killed by signal: SIG]`,或在后端既未提供退出码也未提供信号时追加 `[shell exited]`;随后重置 shell并告知模型下次调用从新 shell 开始。长输出保留仍可读取的最早前缀并追加截断提示;若 PTY 已丢弃真正的开头,结果会明确说明,而不是把尾部伪装成完整输出。超时返回有界的部分输出、关闭状态不确定的 shell并报告该重置。
#### Token 影响
随数据变化。`maxOutputChars` 限制保留的命令输出;固定的截断、前缀丢失、状态、超时与重置诊断可能使结果更长。
#### KV Cache 影响
工具结果以追加方式位于可复用请求前缀之后。
## 已知限制与延后工作
- 工具需要拥有它的 Agent 和真实 PTY 后端。
- 显式 `exit` 与超时会丢弃 shell 状态。取消同样会重置 shell 并丢弃结果,即使已经能观察到完整状态标记也是如此;下次调用创建新 shell。
- 网络访问、软件包镜像等环境事实应写入配置的 `description`,而非包默认描述。

View File

@@ -0,0 +1,55 @@
{
"name": "@deepseek-ai/dsh-tool-bash-persistent",
"description": "Model-facing owner-scoped persistent Bash tool backed by the Harness PTY service",
"version": "0.0.1",
"private": true,
"type": "module",
"main": "lib/index.js",
"types": "lib/types/index.d.ts",
"exports": {
".": {
"types": "./lib/types/index.d.ts",
"default": "./lib/index.js"
},
"./invariant": {
"types": "./lib/types/invariant.d.ts",
"default": "./lib/invariant.js"
},
"./package.json": "./package.json"
},
"files": [
"lib/index.js",
"lib/invariant.js",
"lib/types/**/*.d.ts",
"lib/types/**/*.d.ts.map",
"src"
],
"license": "BSD-3-Clause",
"peerDependencies": {
"@deepseek-ai/dsh-agent": "^0.0.1",
"@deepseek-ai/dsh-invariants": "^0.0.1",
"@deepseek-ai/dsh-pty": "^0.0.1",
"@deepseek-ai/dsh-timeout": "^0.0.1",
"@deepseek-ai/dsh-tools": "^0.0.1",
"cordis": "^4.0.0-rc.7"
},
"dependencies": {
"schemastery": "^3.18.0"
},
"devDependencies": {
"@cordisjs/plugin-include": "workspace:^",
"@cordisjs/plugin-loader": "workspace:^",
"@deepseek-ai/dsh-agent": "workspace:^",
"@deepseek-ai/dsh-invariants": "workspace:^",
"@deepseek-ai/dsh-llm": "workspace:^",
"@deepseek-ai/dsh-pty": "workspace:^",
"@deepseek-ai/dsh-pty-local": "workspace:^",
"@deepseek-ai/dsh-sandbox": "workspace:^",
"@deepseek-ai/dsh-sandbox-policy": "workspace:^",
"@deepseek-ai/dsh-session": "workspace:^",
"@deepseek-ai/dsh-system-prompt": "workspace:^",
"@deepseek-ai/dsh-timeout": "workspace:^",
"@deepseek-ai/dsh-tools": "workspace:^",
"cordis": "^4.0.0-rc.7"
}
}

View File

@@ -0,0 +1,445 @@
/**
* Model-facing persistent `bash` tool over the owner-scoped PTY seam.
* @module @deepseek-ai/dsh-tool-bash-persistent
*/
import { randomUUID } from 'node:crypto'
import type { Context } from 'cordis'
import z from 'schemastery'
import type { Agent } from '@deepseek-ai/dsh-agent'
import type { PtyReadResult, PtySendResult, PtySessionId } from '@deepseek-ai/dsh-pty'
import { deadline, timeoutOf } from '@deepseek-ai/dsh-timeout'
import { defineTool } from '@deepseek-ai/dsh-tools'
// TODO: Replace the file-search advice; arbitrary command output need not come from a searchable file.
const TRUNCATED_MESSAGE = '<response clipped><NOTE>To save on context only part of this file has been shown to you. You should retry this tool after you have searched inside the file with `grep -n` in order to find the line numbers of what you are looking for.</NOTE>'
const LOST_PREFIX_MESSAGE = '<response clipped><NOTE>The beginning of this command output was dropped by the terminal scrollback limit. The following text is the earliest retained output.</NOTE>\n'
const SHELL_RESET_MESSAGE = 'The persistent bash shell was reset; the next bash call starts from the workspace with a fresh current directory and environment.'
const SHELL_PROMPT = '__DSH_PERSISTENT_BASH_PROMPT__ '
const TIMEOUT_CODE = 'PERSISTENT_BASH_TIMEOUT'
// One page is enough to find a just-emitted completion marker; the full
// scrollback is assembled only when a command settles or needs partial output.
const SCROLLBACK_PAGE_LINES = 1_000
const POLL_INTERVAL_MS = 25
const DEFAULT_DESCRIPTION = 'Run commands in a persistent bash shell. State, including the current directory and exported environment variables, persists across calls for this agent.'
interface ResolvedConfig {
backendType: string
timeoutMs: number
maxOutputChars: number
description: string
}
interface CommandMarkers {
start: string
end: string
}
interface RetainedOutput {
text: string
truncated: boolean
}
interface CapturedOutput {
text: string
incomplete: boolean
exitCode?: number
}
interface PersistentShells {
get(owner: Agent, signal: AbortSignal): Promise<PtySessionId>
reset(owner: Agent, reason: string): Promise<void>
}
function maybeTruncate(content: string, maxOutputChars: number, incomplete = false): string {
if (content.length <= maxOutputChars && !incomplete) return content
return content.length <= maxOutputChars
? content + TRUNCATED_MESSAGE
: content.slice(0, maxOutputChars) + TRUNCATED_MESSAGE
}
function markers(): CommandMarkers {
const nonce = randomUUID()
return {
start: `__DSH_PERSISTENT_BASH_START_${nonce}__`,
end: `__DSH_PERSISTENT_BASH_END_${nonce}:`,
}
}
function quoteForBash(value: string): string {
return `$'${value
.replaceAll('\\', '\\\\')
.replaceAll("'", "\\'")
.replaceAll('\r', '\\r')
.replaceAll('\n', '\\n')}'`
}
function wrapCommand(command: string, marker: CommandMarkers): string {
// Keep the wrapper on one physical line. An interactive bash prints PS2 for
// embedded newlines before executing the buffer, which would leak terminal
// prompts and marker source text into the model-facing result.
return `printf '%s\\n' ${quoteForBash(marker.start)}; eval -- ${quoteForBash(command)}; __dsh_persistent_bash_status=$?; printf '%s%s\\n' ${quoteForBash(marker.end)} "$__dsh_persistent_bash_status"`
}
function stripPrompt(text: string): string {
let result = text.replace(/\r?\n$/, '')
while (result.endsWith(SHELL_PROMPT)) {
result = result.slice(0, -SHELL_PROMPT.length)
}
return result.endsWith('\n') ? result.slice(0, -1) : result
}
function commandOutput(
snapshot: RetainedOutput,
marker: CommandMarkers,
): CapturedOutput | undefined {
const text = snapshot.text
const end = text.lastIndexOf(marker.end)
const status = /^(\d+)\r?\n/.exec(text.slice(end + marker.end.length))?.[1]
if (status === undefined) return undefined
const startMarker = text.lastIndexOf(marker.start, end)
const start = startMarker < 0 ? 0 : startMarker + marker.start.length
return {
text: stripPrompt(text.slice(start, end).replace(/^\r?\n/, '')),
incomplete: startMarker < 0,
exitCode: Number(status),
}
}
function promptCompleted(result: PtySendResult): boolean {
return result.viewport.endsWith(SHELL_PROMPT)
|| result.viewport.endsWith(`${SHELL_PROMPT}\r\n`)
|| result.viewport.endsWith(`${SHELL_PROMPT}\n`)
}
function partialOutput(
snapshot: RetainedOutput,
marker: CommandMarkers,
fallback: string,
fallbackTruncated = false,
): CapturedOutput {
const startMarker = snapshot.text.lastIndexOf(marker.start)
if (startMarker >= 0) {
return {
text: stripPrompt(snapshot.text.slice(startMarker + marker.start.length).replace(/^\r?\n/, '')),
incomplete: false,
}
}
const fallbackStart = fallback.lastIndexOf(marker.start)
const afterStart = fallbackStart < 0
? fallback
: fallback.slice(fallbackStart + marker.start.length).replace(/^\r?\n/, '')
const fallbackEnd = afterStart.lastIndexOf(marker.end)
const beforeEnd = fallbackEnd < 0 ? afterStart : afterStart.slice(0, fallbackEnd)
return {
text: stripPrompt(beforeEnd.replaceAll(SHELL_PROMPT, '')),
incomplete: fallbackTruncated || fallbackStart < 0,
}
}
async function pause(): Promise<void> {
await new Promise(resolve => setTimeout(resolve, POLL_INTERVAL_MS))
}
function nextScrollbackOffset(page: PtyReadResult, offset: number): number | undefined {
if (page.text.length === 0 || page.lineEnd <= offset) return undefined
return page.lineEnd
}
function retainedScrollback(
ctx: Context,
owner: Agent,
id: PtySessionId,
latest = ctx.pty.read(owner, id, { offset: 0, count: SCROLLBACK_PAGE_LINES }),
): RetainedOutput {
const pages: string[] = latest.text.length === 0 ? [] : [latest.text]
let offset = latest.lineEnd
let truncated = latest.truncated
while (true) {
if (offset >= latest.totalLines) break
const page = ctx.pty.read(owner, id, { offset, count: SCROLLBACK_PAGE_LINES })
truncated ||= page.truncated
if (page.text.length > 0) pages.unshift(page.text)
const next = nextScrollbackOffset(page, offset)
if (next === undefined || next >= page.totalLines) break
offset = next
}
return { text: pages.join('\n'), truncated }
}
function renderCaptured(output: CapturedOutput, maxOutputChars: number): string {
const rendered = maybeTruncate(output.text, maxOutputChars, output.incomplete)
const withPrefix = output.incomplete && output.text.length > 0
? LOST_PREFIX_MESSAGE + rendered
: rendered
const marker = output.exitCode !== undefined && output.exitCode !== 0
? `[exit code: ${output.exitCode}]`
: undefined
return appendStatusMarker(withPrefix, marker)
}
function appendStatusMarker(content: string, marker: string | undefined): string {
if (marker === undefined) return content
return content.length === 0 ? marker : `${content}\n${marker}`
}
function renderShellExitStatus(
content: string,
exitCode: number | null,
signal: NodeJS.Signals | null,
): string {
const marker = signal !== null
? `[shell killed by signal: ${signal}]`
: exitCode !== null
? `[shell exited: code ${exitCode}]`
: '[shell exited]'
return appendStatusMarker(content, marker)
}
function persistentShells(ctx: Context, config: ResolvedConfig): PersistentShells {
const pending = new WeakMap<Agent, Promise<PtySessionId>>()
const live = new Map<Agent, PtySessionId>()
const creating = new Set<Promise<PtySessionId>>()
const ownerCleanupInstalled = new WeakSet<Agent>()
const lifecycle = new AbortController()
const close = async (owner: Agent, id: PtySessionId, reason: string): Promise<void> => {
if (!ctx.pty.list(owner).some(snapshot => snapshot.sessionId === id)) return
await ctx.pty.kill(owner, id, reason)
}
ctx.effect(() => async () => {
lifecycle.abort(new Error('tool-bash-persistent disposed during shell creation'))
await Promise.allSettled([...creating])
const closing = [...live].map(async ([owner, id]) => { await close(owner, id, 'tool-bash-persistent disposed') })
await Promise.all(closing)
live.clear()
}, 'tool-bash-persistent shell cleanup')
const reset = async (owner: Agent, reason: string): Promise<void> => {
pending.delete(owner)
const id = live.get(owner)
live.delete(owner)
if (id !== undefined) await close(owner, id, reason)
}
const get = (owner: Agent, signal: AbortSignal): Promise<PtySessionId> => {
const existing = pending.get(owner)
if (existing !== undefined) return existing
const combinedSignal = AbortSignal.any([signal, lifecycle.signal])
const creation = (async () => {
try {
const cwd = owner.session.header.cwd
const spawned = await ctx.pty.spawn(owner, {
type: config.backendType,
...cwd === undefined ? {} : { cwd },
}, combinedSignal)
live.set(owner, spawned.sessionId)
if (!ownerCleanupInstalled.has(owner)) {
ownerCleanupInstalled.add(owner)
owner.ctx.effect(() => () => {
pending.delete(owner)
live.delete(owner)
}, 'tool-bash-persistent owner cache cleanup')
}
const setup = ctx.pty.startSend(owner, spawned.sessionId, {
text: `stty -echo; PS1=${quoteForBash(SHELL_PROMPT)}`,
submit: true,
signal: combinedSignal,
})
const result = await setup.done
if (result.sessionStatus.kind === 'exited' || result.waitReason === 'timeout') {
throw new Error('persistent bash shell did not accept initialization')
}
return spawned.sessionId
} catch (error: unknown) {
await reset(owner, 'persistent bash initialization failed')
throw error
}
})()
const tracked = creation.finally(() => {
creating.delete(tracked)
})
creating.add(tracked)
pending.set(owner, tracked)
return tracked
}
return { get, reset }
}
async function executeCommand(
ctx: Context,
shells: PersistentShells,
owner: Agent,
command: string,
config: ResolvedConfig,
upstream: AbortSignal,
): Promise<string> {
using commandDeadline = deadline(upstream, config.timeoutMs, TIMEOUT_CODE)
const id = await shells.get(owner, commandDeadline.signal)
const marker = markers()
const wrapped = wrapCommand(command, marker)
let first = true
let fallback = ''
let fallbackTruncated = false
while (true) {
let operation
let result
try {
operation = ctx.pty.startSend(owner, id, {
text: first ? wrapped : '',
submit: first,
signal: commandDeadline.signal,
})
first = false
result = await operation.done
} catch (error: unknown) {
await shells.reset(owner, 'persistent bash send failed')
throw error
}
const incremental = operation.readOutput()
fallback = incremental.delta.length > 0 ? fallback + incremental.delta : result.viewport
fallbackTruncated ||= incremental.truncated || result.truncated
const latest = ctx.pty.read(owner, id, { offset: 0, count: SCROLLBACK_PAGE_LINES })
const timedOut = timeoutOf(commandDeadline.signal, TIMEOUT_CODE)
if (timedOut !== undefined) {
const snapshot = retainedScrollback(ctx, owner, id, latest)
const partial = renderCaptured(
partialOutput(snapshot, marker, fallback, fallbackTruncated),
config.maxOutputChars,
)
await shells.reset(owner, 'persistent bash command timed out')
return [
// TODO: Report a timeout only; this signal does not establish an OOM.
`Your command timed out after ${Math.round(timedOut.timeoutMs / 1000)} seconds or experienced an OOM error. Below is partial output:`,
partial,
SHELL_RESET_MESSAGE,
].join('\n')
}
if (commandDeadline.signal.aborted) {
await shells.reset(owner, 'persistent bash command aborted')
commandDeadline.signal.throwIfAborted()
}
if (latest.text.includes(marker.end)) {
const complete = commandOutput(retainedScrollback(ctx, owner, id, latest), marker)
if (complete !== undefined) return renderCaptured(complete, config.maxOutputChars)
}
if (result.sessionStatus.kind === 'exited') {
const snapshot = retainedScrollback(ctx, owner, id, latest)
await shells.reset(owner, 'persistent bash shell exited')
return [
renderShellExitStatus(
renderCaptured(partialOutput(snapshot, marker, fallback, fallbackTruncated), config.maxOutputChars),
result.sessionStatus.exitCode,
result.sessionStatus.signal,
),
SHELL_RESET_MESSAGE,
].filter(part => part.length > 0).join('\n')
}
if (promptCompleted(result)) {
const snapshot = retainedScrollback(ctx, owner, id, latest)
return renderCaptured(
partialOutput(snapshot, marker, fallback, fallbackTruncated),
config.maxOutputChars,
)
}
await pause()
}
}
/**
* Register the model-facing persistent `bash` tool.
* @param ctx - plugin context carrying tools and the owner-scoped PTY service.
* @param config - selected PTY backend and command deadline.
*/
function registerPersistentBash(ctx: Context, config: ResolvedConfig): void {
const shells = persistentShells(ctx, config)
const queues = new WeakMap<Agent, Promise<void>>()
const serialized = async <T>(owner: Agent, operation: () => Promise<T>): Promise<T> => {
const prior = queues.get(owner) ?? Promise.resolve()
const run = prior.then(operation, operation)
const tail = run.then(() => undefined, () => undefined)
queues.set(owner, tail)
try {
return await run
} finally {
if (queues.get(owner) === tail) queues.delete(owner)
}
}
ctx.tools.register(defineTool({
name: 'bash',
description: config.description,
parameters: {
command: {
type: 'string',
required: true,
description: 'The bash command to run. Relative path is preferred in the command.',
},
},
output: {
schema: { type: 'string' },
render: (_args, value) => [{ type: 'text', text: value }],
},
async execute(args, exec) {
if (args.command.trim().length === 0) throw new Error('command must be a non-empty string')
const owner = exec.agent
if (owner === undefined) throw new Error('bash requires an owning agent session')
return serialized(owner, async () => {
exec.signal.throwIfAborted()
return executeCommand(ctx, shells, owner, args.command, config, exec.signal)
})
},
presentCall: args => ({ card: 'terminal', title: args.command }),
}))
}
export const name = 'tool-bash-persistent'
export const inject = ['tools', 'pty']
/** Configuration for the persistent Bash tool. */
export interface Config {
/** PTY backend used for each owner-isolated persistent shell (default `shell`). */
backendType?: string
/** Wall-clock limit for one command (default 300000). */
timeoutMs?: number
/** Maximum returned command-output characters before clipping (default 16000). */
maxOutputChars?: number
/** Model-facing tool description; deployments may describe their environment. */
description?: string
}
/** Runtime configuration schema for the persistent Bash tool. */
export const Config: z<Config> = z.object({
backendType: z.string().default('shell'),
timeoutMs: z.number().default(300_000),
maxOutputChars: z.number().default(16_000),
description: z.string().default(DEFAULT_DESCRIPTION),
})
/** Register one owner-scoped persistent `bash` tool. */
export function apply(ctx: Context, config: Config): void {
const resolved: ResolvedConfig = {
backendType: config.backendType ?? 'shell',
timeoutMs: config.timeoutMs ?? 300_000,
maxOutputChars: config.maxOutputChars ?? 16_000,
description: config.description ?? DEFAULT_DESCRIPTION,
}
if (resolved.backendType.trim().length === 0) {
throw new Error('tool-bash-persistent: backendType must be non-empty')
}
if (!Number.isSafeInteger(resolved.timeoutMs) || resolved.timeoutMs <= 0) {
throw new Error('tool-bash-persistent: timeoutMs must be a positive safe integer')
}
if (!Number.isSafeInteger(resolved.maxOutputChars) || resolved.maxOutputChars <= 0) {
throw new Error('tool-bash-persistent: maxOutputChars must be a positive safe integer')
}
if (resolved.description.trim().length === 0) {
throw new Error('tool-bash-persistent: description must be non-empty')
}
registerPersistentBash(ctx, resolved)
}

View File

@@ -0,0 +1,31 @@
/**
* Package-owned invariant companion for `@deepseek-ai/dsh-tool-bash-persistent`.
* @module @deepseek-ai/dsh-tool-bash-persistent/invariant
*/
/* jscpd:ignore-start */
import type { Context } from 'cordis'
import type { InvariantInstaller } from '@deepseek-ai/dsh-invariants'
const PACKAGE_NAME = '@deepseek-ai/dsh-tool-bash-persistent'
/** Cordis companion plugin name. */
export const name = 'tool-bash-persistent-invariant'
/** Service required before the companion can reserve package ownership. */
export const inject = ['invariants']
/**
* No runtime invariant: the adapter's private owner-to-shell cache has no
* observable event or data relation. Lifecycle tests prove its cleanup without
* adding a public surface solely for an invariant.
*/
const install: InvariantInstaller = () => {}
/**
* Register this package's invariant companion.
* @param ctx - Cordis context carrying the invariant service.
* @returns the installed registration's disposer after setup succeeds.
*/
export const apply = (ctx: Context): Promise<() => void> =>
Promise.resolve(ctx.invariants.register(PACKAGE_NAME, install))
/* jscpd:ignore-end */

View File

@@ -0,0 +1,157 @@
import { mkdtemp, rm, writeFile } from 'node:fs/promises'
import { tmpdir } from 'node:os'
import { join } from 'node:path'
import { pathToFileURL } from 'node:url'
import { afterEach, describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import Loader from '@cordisjs/plugin-loader'
import Include from '@cordisjs/plugin-include'
import { CallId } from '@deepseek-ai/dsh-llm'
import { Session, SessionId } from '@deepseek-ai/dsh-session'
import AgentRegistry from '@deepseek-ai/dsh-agent'
import type { Agent } from '@deepseek-ai/dsh-agent'
import PtyService from '@deepseek-ai/dsh-pty'
import * as PtyLocal from '@deepseek-ai/dsh-pty-local'
import SandboxProvider from '@deepseek-ai/dsh-sandbox'
import type { ConfinedArgv, SandboxPolicy } from '@deepseek-ai/dsh-sandbox'
import SandboxPolicyService from '@deepseek-ai/dsh-sandbox-policy'
import SystemPrompt from '@deepseek-ai/dsh-system-prompt'
import ToolRegistry from '@deepseek-ai/dsh-tools'
import * as ToolBashPersistent from '@deepseek-ai/dsh-tool-bash-persistent'
let root: string | undefined
let context: Context | undefined
afterEach(async () => {
await context?.fiber.dispose()
context = undefined
if (root !== undefined) await rm(root, { recursive: true, force: true })
root = undefined
})
class PassthroughSandbox extends SandboxProvider {
confine(argv: readonly string[], _policy: SandboxPolicy): ConfinedArgv {
return { argv: [...argv], enforcement: 'full', denialSignatures: [], runnerFailureSignatures: [] }
}
}
function agent(ctx: Context, cwd: string): Agent {
const id = SessionId('persistent-bash-loader-agent')
const scope = ctx.plugin(() => {})
const value: Agent = {
id,
options: {},
session: new Session(id, [], { version: 0, id, createdAt: 0, cwd }),
status: 'idle',
acceptsNextStep: false,
ctx: scope.ctx,
followup: () => {},
steer: () => {},
inject: () => {},
send: () => {},
updateInbox: () => 'not-found',
cancel() {},
whenIdle: () => Promise.resolve(),
}
ctx.agents.register(value)
return value
}
function text(result: { content: { type: string; text?: string }[] }): string {
return result.content.filter(block => block.type === 'text').map(block => block.text).join('')
}
const suite = process.platform === 'linux' || process.platform === 'darwin' ? describe : describe.skip
suite('persistent Bash through a real cordis.yml Loader composition', () => {
it('preserves cwd and environment across calls', async () => {
root = await mkdtemp(join(tmpdir(), 'dsh-persistent-bash-loader-'))
const configPath = join(root, 'cordis.yml')
await writeFile(configPath, [
"- name: '@deepseek-ai/dsh-agent'",
"- name: '@deepseek-ai/dsh-system-prompt'",
"- name: '@deepseek-ai/dsh-tools'",
"- name: '@deepseek-ai/dsh-pty'",
"- name: '@deepseek-ai/dsh-test-sandbox'",
"- name: '@deepseek-ai/dsh-sandbox-policy'",
' config:',
' mode: danger-full-access',
` workspaceRoot: ${JSON.stringify(root)}`,
"- name: '@deepseek-ai/dsh-pty-local'",
' config:',
' pollIntervalMs: 10',
' exactProbeAfterMs: 20',
' idleSilenceMs: 100',
' handoffGraceMs: 100',
' scrollbackLines: 20000',
' timeoutMs: 2000',
' disposeGraceMs: 500',
"- name: '@deepseek-ai/dsh-tool-bash-persistent'",
' config:',
' timeoutMs: 5000',
'',
].join('\n'))
context = new Context()
context.baseUrl = pathToFileURL(root).href + '/'
await context.plugin(Loader)
context.loader.builtins.include = Include
const modules = new Map<string, unknown>([
['@deepseek-ai/dsh-agent', AgentRegistry],
['@deepseek-ai/dsh-system-prompt', SystemPrompt],
['@deepseek-ai/dsh-tools', ToolRegistry],
['@deepseek-ai/dsh-pty', PtyService],
['@deepseek-ai/dsh-test-sandbox', PassthroughSandbox],
['@deepseek-ai/dsh-sandbox-policy', SandboxPolicyService],
['@deepseek-ai/dsh-pty-local', PtyLocal],
['@deepseek-ai/dsh-tool-bash-persistent', ToolBashPersistent],
])
context.loader.internal = {
version: 'v2',
async import(specifier: string) {
if (!modules.has(specifier)) throw new Error(`unexpected Loader import: ${specifier}`)
return modules.get(specifier)
},
} as unknown as NonNullable<typeof context.loader.internal>
await context.loader.create({ name: 'cordis:include', config: { path: pathToFileURL(configPath).href } })
await context.loader.await()
const owner = agent(context, root)
const signal = new AbortController().signal
const execute = (id: string, command: string) => context!.tools.execute({
signal,
callId: CallId(id),
name: 'bash',
arguments: { command },
agent: owner,
})
expect(context.tools.schemas().map(schema => schema.name)).toEqual(['bash'])
await execute('state', 'export KEEP=loader; mkdir -p nested; cd nested')
const observed = text(await execute('observe', 'printf "cwd=%s keep=%s\\n" "$PWD" "$KEEP"'))
expect(observed).toContain(`cwd=${join(root, 'nested')} keep=loader`)
expect(observed).not.toContain('DSH_PERSISTENT_BASH')
const multiline = text(await execute(
'multiline',
'value="line one"\nprintf "%s:%s\\n" "$value" "it\'s fine"',
))
expect(multiline).toBe("line one:it's fine")
expect(multiline).not.toContain('DSH_PERSISTENT_BASH')
const heredoc = text(await execute(
'heredoc',
"cat <<'EOF'\nalpha\nbeta\nEOF",
))
expect(heredoc).toBe('alpha\nbeta')
const large = text(await execute('large-output', 'seq 1 12050'))
expect(large.startsWith('1\n2\n3\n')).toBe(true)
expect(large).toContain('<response clipped>')
expect(large).not.toContain('beginning of this command output was dropped')
const exited = text(await execute('exit', 'exit'))
expect(exited).toContain('next bash call starts from the workspace')
expect(text(await execute('after-exit', 'printf "%s\\n" "$PWD"'))).toBe(root)
}, 20_000)
})

View File

@@ -0,0 +1,538 @@
import { afterEach, describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import { CallId } from '@deepseek-ai/dsh-llm'
import { Session, SessionId } from '@deepseek-ai/dsh-session'
import AgentRegistry from '@deepseek-ai/dsh-agent'
import type { Agent } from '@deepseek-ai/dsh-agent'
import PtyService from '@deepseek-ai/dsh-pty'
import type {
PtyBackend,
PtyBackendSession,
PtyReadRequest,
PtySendOperation,
PtySendRequest,
PtySessionStatus,
PtySignal,
PtyWaitReason,
} from '@deepseek-ai/dsh-pty'
import SystemPrompt from '@deepseek-ai/dsh-system-prompt'
import ToolRegistry from '@deepseek-ai/dsh-tools'
import * as ToolBashPersistent from '@deepseek-ai/dsh-tool-bash-persistent'
const contexts: Context[] = []
let callNumber = 0
afterEach(async () => {
for (const ctx of contexts.splice(0)) await ctx.fiber.dispose()
})
function agent(ctx: Context, cwd: string | undefined): Agent {
const id = SessionId(`persistent-bash-owner-${callNumber}`)
const scope = ctx.plugin(() => {})
const value: Agent = {
id,
options: {},
session: new Session(id, [], {
version: 0,
id,
createdAt: 0,
...cwd === undefined ? {} : { cwd },
}),
status: 'idle',
acceptsNextStep: false,
ctx: scope.ctx,
followup: () => {},
steer: () => {},
inject: () => {},
send: () => {},
updateInbox: () => 'not-found',
cancel() {},
whenIdle: () => Promise.resolve(),
}
ctx.agents.register(value)
return value
}
function text(result: { content: { type: string; text?: string }[] }): string {
return result.content.filter(block => block.type === 'text').map(block => block.text).join('')
}
function call(
ctx: Context,
owner: Agent | undefined,
command: string,
signal = new AbortController().signal,
) {
return ctx.tools.execute({
signal,
callId: CallId(`persistent-bash-${++callNumber}`),
name: 'bash',
arguments: { command },
...owner === undefined ? {} : { agent: owner },
})
}
type StubMode =
| 'normal'
| 'prompt-only'
| 'prompt-crlf'
| 'empty-read'
| 'stalled-read'
| 'exit'
| 'signal-exit'
| 'unknown-exit'
| 'wait-for-abort'
| 'end-on-abort'
| 'idle-then-normal'
| 'large'
| 'nonzero'
| 'torn-status'
| 'finish-torn-status'
| 'end-only'
| 'init-exit'
| 'init-timeout'
| 'spawn-error'
| 'send-error'
| 'prompt-after-idle'
| 'empty-page-after-latest'
class StubPtySession implements PtyBackendSession {
readonly motd = '__DSH_PERSISTENT_BASH_PROMPT__ '
readonly pid = 123
statusValue: PtySessionStatus = { kind: 'running' }
scrollback = this.motd
closed: string[] = []
mode: StubMode
sends = 0
pendingText = ''
historyTruncated = false
constructor(mode: StubMode) {
this.mode = mode
}
startSend(request: PtySendRequest): PtySendOperation {
this.sends += 1
if (request.text.startsWith('stty -echo')) {
if (this.mode === 'init-exit') {
this.statusValue = { kind: 'exited', exitCode: 1, signal: null }
return this.operation(Promise.resolve(this.result('', 'session_exit')))
}
if (this.mode === 'init-timeout') {
return this.operation(Promise.resolve(this.result('', 'timeout')))
}
return this.operation(Promise.resolve(this.result(this.motd, 'stdin_read')))
}
if (this.mode === 'send-error') throw new Error('stub send failed')
if (this.mode === 'wait-for-abort' || this.mode === 'end-on-abort') {
const done = new Promise<ReturnType<StubPtySession['result']>>((resolve) => {
request.signal?.addEventListener('abort', () => {
const start = /__DSH_PERSISTENT_BASH_START_[^_]+(?:-[^_]+)*__/.exec(request.text)?.[0]
const end = /__DSH_PERSISTENT_BASH_END_[^:]+:/.exec(request.text)?.[0]
const output = this.mode === 'end-on-abort'
? `${start ?? ''}\ninterrupted\n${end ?? ''}130\n${this.motd}`
: 'partial output'
this.scrollback += output
resolve(this.result(output, 'stdin_read'))
}, { once: true })
})
return this.operation(done)
}
if (this.mode === 'idle-then-normal') {
this.mode = 'normal'
this.pendingText = request.text
return this.operation(Promise.resolve(this.result('', 'inferred_idle')))
}
if (this.mode === 'prompt-after-idle') {
if (request.text.length > 0) {
const start = /__DSH_PERSISTENT_BASH_START_[^_]+(?:-[^_]+)*__/.exec(request.text)?.[0]
const output = `${start ?? ''}\npartial syntax output\n`
this.scrollback += output
return this.operation(Promise.resolve(this.result(output, 'inferred_idle')))
}
const output = `bash: syntax error\n${this.motd}`
this.scrollback += output
return this.operation(Promise.resolve(this.result(output, 'stdin_read')))
}
if (this.mode === 'prompt-only' || this.mode === 'prompt-crlf') {
const newline = this.mode === 'prompt-crlf' ? '\r\n' : '\n'
const output = `bash: syntax error${newline}${this.motd}${newline}`
this.scrollback += output
return this.operation(Promise.resolve(this.result(output, 'stdin_read')))
}
const sent = request.text.length > 0 ? request.text : this.pendingText
this.pendingText = ''
const start = /__DSH_PERSISTENT_BASH_START_[^_]+(?:-[^_]+)*__/.exec(sent)?.[0]
const end = /__DSH_PERSISTENT_BASH_END_[^:]+:/.exec(sent)?.[0]
if (this.mode === 'torn-status') {
const output = `${start ?? ''}\nhello from stub\n${end ?? ''}`
this.scrollback += output
this.mode = 'finish-torn-status'
return this.operation(Promise.resolve(this.result(output, 'inferred_idle')))
}
if (this.mode === 'finish-torn-status') {
const output = `7\n${this.motd}`
this.scrollback += output
return this.operation(Promise.resolve(this.result(output, 'stdin_read')))
}
if (this.mode === 'end-only') {
const output = `recovered output\n${end ?? ''}0\n${this.motd}`
this.scrollback += output
return this.operation(Promise.resolve(this.result(output, 'stdin_read')))
}
const commandOutput = this.mode === 'large'
? 'x'.repeat(100)
: this.mode === 'nonzero' ? '' : 'hello from stub'
const exitCode = this.mode === 'nonzero' ? 7 : 0
const output = `${start ?? ''}\n${commandOutput}\n${end ?? ''}${exitCode}\n${this.motd}`
this.scrollback += output
if (this.mode === 'exit' || this.mode === 'signal-exit' || this.mode === 'unknown-exit') {
const exitedOutput = `${start ?? ''}\nhello from stub\n`
this.scrollback = this.scrollback.slice(0, -output.length) + exitedOutput
this.statusValue = this.mode === 'signal-exit'
? { kind: 'exited', exitCode: null, signal: 'SIGTERM' }
: this.mode === 'exit'
? { kind: 'exited', exitCode: 9, signal: null }
: { kind: 'exited', exitCode: null, signal: null }
return this.operation(Promise.resolve(this.result(exitedOutput, 'session_exit')))
}
return this.operation(Promise.resolve(this.result(output, 'stdin_read')))
}
read(request: PtyReadRequest) {
if (this.mode === 'empty-read') {
return { text: '', totalLines: 0, lineBegin: 0, lineEnd: 0, truncated: false }
}
if (this.mode === 'stalled-read') {
return { text: 'stalled', totalLines: 1, lineBegin: 0, lineEnd: 0, truncated: false }
}
if (this.mode === 'empty-page-after-latest' && (request.offset ?? 0) > 0) {
return { text: '', totalLines: 2, lineBegin: 1, lineEnd: 1, truncated: false }
}
const lines = this.scrollback.split('\n')
return {
text: this.scrollback,
totalLines: this.mode === 'empty-page-after-latest' ? lines.length + 1 : lines.length,
lineBegin: 0,
lineEnd: this.mode === 'empty-page-after-latest' ? 1 : lines.length,
truncated: this.historyTruncated,
}
}
signal(_signal: PtySignal) {
return Promise.resolve({ delivered: true as const, targetPgid: 123 })
}
status() {
return this.statusValue
}
async close(reason: string) {
this.closed.push(reason)
this.statusValue = { kind: 'exited', exitCode: 0, signal: null }
}
private result(viewport: string, waitReason: PtyWaitReason) {
return { viewport, waitReason, sessionStatus: this.statusValue, truncated: false }
}
private operation(done: Promise<ReturnType<StubPtySession['result']>>): PtySendOperation {
return {
done,
readOutput: () => ({ delta: '', truncated: false }),
cancel: () => false,
}
}
}
function stubBackend(initialMode: StubMode = 'normal') {
const sessions: StubPtySession[] = []
const backend: PtyBackend = {
type: 'stub',
async spawn() {
if (initialMode === 'spawn-error') throw new Error('stub spawn failed')
const session = new StubPtySession(initialMode)
sessions.push(session)
return session
},
}
return { backend, sessions }
}
async function setup(
config: ToolBashPersistent.Config = { backendType: 'stub' },
initialMode: StubMode = 'normal',
) {
const ctx = new Context()
contexts.push(ctx)
await ctx.plugin(SystemPrompt)
await ctx.plugin(ToolRegistry)
await ctx.plugin(AgentRegistry)
await ctx.plugin(PtyService)
const stub = stubBackend(initialMode)
ctx.pty.registerBackend(stub.backend)
const fiber = await ctx.plugin(ToolBashPersistent, config)
return { ctx, stub, fiber, owner: agent(ctx, '/workspace') }
}
describe('tool-bash-persistent', () => {
it('registers a configurable schema and reuses one owner shell', async () => {
const { ctx, owner, stub, fiber } = await setup({
backendType: 'stub',
description: 'deployment-specific persistent shell',
})
const schema = ctx.tools.schemas()[0]
expect(ctx.tools.schemas().map(item => item.name)).toEqual(['bash'])
expect(schema?.description).toBe('deployment-specific persistent shell')
expect(schema?.parameters).toMatchObject({
required: ['command'],
properties: { command: { type: 'string' } },
})
expect(ctx.tools.get('bash')?.presentCall?.({ command: 'pwd' }))
.toEqual({ card: 'terminal', title: 'pwd' })
expect(text(await call(ctx, owner, 'echo one'))).toBe('hello from stub')
expect(text(await call(ctx, owner, 'echo two'))).toBe('hello from stub')
expect(stub.sessions).toHaveLength(1)
expect(stub.sessions[0]?.sends).toBe(3)
const ownerWithoutCwd = agent(ctx, undefined)
expect(text(await call(ctx, ownerWithoutCwd, 'pwd'))).toBe('hello from stub')
expect(stub.sessions).toHaveLength(2)
await fiber.dispose()
expect(ctx.tools.schemas()).toEqual([])
expect(ctx.tools.get('bash')).toBeUndefined()
})
it('handles inferred idle, prompt fallback, shell exit, clipping, and cleanup', async () => {
const { ctx, owner, stub, fiber } = await setup({
backendType: 'stub',
maxOutputChars: 10,
})
await call(ctx, owner, 'warm up')
const session = stub.sessions[0]!
session.mode = 'idle-then-normal'
expect(text(await call(ctx, owner, 'silent then complete'))).toContain('hello from')
session.mode = 'prompt-only'
const promptFallback = text(await call(ctx, owner, 'bad {'))
expect(promptFallback).toContain('bash: synt')
expect(promptFallback).not.toContain('DSH_PERSISTENT_BASH_PROMPT')
session.mode = 'prompt-crlf'
session.scrollback = ''
const crlfPromptFallback = text(await call(ctx, owner, 'bad {'))
expect(crlfPromptFallback).toContain('bash: synt')
expect(crlfPromptFallback).not.toContain('DSH_PERSISTENT_BASH_PROMPT')
session.mode = 'end-only'
session.scrollback = ''
const missingStart = text(await call(ctx, owner, 'recover marker'))
expect(missingStart).toContain('recovered')
expect(missingStart).toContain('beginning of this command output was dropped')
expect(missingStart).toContain('<response clipped>')
session.mode = 'large'
expect(text(await call(ctx, owner, 'large'))).toContain('<response clipped>')
session.mode = 'nonzero'
expect(text(await call(ctx, owner, 'false'))).toBe('[exit code: 7]')
session.mode = 'exit'
const exited = text(await call(ctx, owner, 'exit'))
expect(exited).toContain('hello from')
expect(exited).toContain('[shell exited: code 9]')
expect(exited).not.toContain('[exit code: 9]')
expect(exited).toContain('next bash call starts from the workspace')
expect(session.closed).toContain('persistent bash shell exited')
await call(ctx, owner, 'new shell')
expect(stub.sessions).toHaveLength(2)
const replacement = stub.sessions[1]!
replacement.mode = 'signal-exit'
expect(text(await call(ctx, owner, 'kill shell')))
.toContain('[shell killed by signal: SIGTERM]')
await call(ctx, owner, 'another shell')
expect(stub.sessions).toHaveLength(3)
const externallyClosed = ctx.pty.list(owner)[0]?.sessionId
expect(externallyClosed).toBeDefined()
await ctx.pty.kill(owner, externallyClosed!, 'external cleanup')
await fiber.dispose()
expect(stub.sessions[2]?.closed).toEqual(['external cleanup'])
})
it('waits for status digits after a torn completion marker', async () => {
const { ctx, owner, stub } = await setup({ backendType: 'stub', maxOutputChars: 1_000 })
await call(ctx, owner, 'warm up')
stub.sessions[0]!.mode = 'torn-status'
stub.sessions[0]!.scrollback = ''
expect(text(await call(ctx, owner, 'torn status'))).toBe('hello from stub\n[exit code: 7]')
})
it('reports a shell exit when the backend has no code or signal', async () => {
const { ctx, owner, stub } = await setup({ backendType: 'stub' })
await call(ctx, owner, 'warm up')
stub.sessions[0]!.mode = 'unknown-exit'
expect(text(await call(ctx, owner, 'exit without status'))).toContain('[shell exited]')
})
it('marks a short missing-prefix result and tolerates exhausted scrollback pages', async () => {
const { ctx, owner, stub } = await setup({ backendType: 'stub', maxOutputChars: 1_000 })
await call(ctx, owner, 'warm up')
const session = stub.sessions[0]!
session.mode = 'end-only'
session.scrollback = ''
expect(text(await call(ctx, owner, 'missing start')))
.toContain('beginning of this command output was dropped')
session.mode = 'empty-read'
expect(text(await call(ctx, owner, 'empty page'))).toContain('hello from stub')
session.mode = 'stalled-read'
expect(text(await call(ctx, owner, 'stalled page'))).toContain('hello from stub')
session.mode = 'empty-page-after-latest'
expect(text(await call(ctx, owner, 'empty continuation page'))).toContain('hello from stub')
})
it('sanitizes a prompt fallback reached after multiple polling rounds', async () => {
const { ctx, owner, stub } = await setup({ backendType: 'stub', maxOutputChars: 1_000 })
await call(ctx, owner, 'warm up')
const session = stub.sessions[0]!
session.mode = 'prompt-after-idle'
session.scrollback = ''
const result = text(await call(ctx, owner, 'bad {'))
expect(result).toContain('partial syntax output')
expect(result).toContain('bash: syntax error')
expect(result).not.toContain('DSH_PERSISTENT_BASH_PROMPT')
expect(result).not.toContain('DSH_PERSISTENT_BASH_START')
})
it('does not attribute old scrollback truncation to a complete current command', async () => {
const { ctx, owner, stub } = await setup({ backendType: 'stub', maxOutputChars: 1_000 })
await call(ctx, owner, 'warm up')
stub.sessions[0]!.historyTruncated = true
const result = text(await call(ctx, owner, 'short command'))
expect(result).toBe('hello from stub')
expect(result).not.toContain('<response clipped>')
expect(result).not.toContain('beginning of this command output was dropped')
})
it('closes a timed-out shell and reports bounded partial output', async () => {
const { ctx, owner, stub } = await setup({ backendType: 'stub', timeoutMs: 10 })
await call(ctx, owner, 'warm up')
stub.sessions[0]!.mode = 'wait-for-abort'
const result = await call(ctx, owner, 'hang')
expect(text(result)).toContain('timed out after 0 seconds or experienced an OOM error')
expect(text(result)).toContain('partial output')
expect(text(result)).toContain('next bash call starts from the workspace')
expect(stub.sessions[0]?.closed).toContain('persistent bash command timed out')
})
it.each(['wait-for-abort', 'end-on-abort'] as const)(
'cancels %s work, resets the shell, and releases a queued call',
async (mode) => {
const { ctx, owner, stub } = await setup({ backendType: 'stub', timeoutMs: 5_000 })
await call(ctx, owner, 'warm up')
stub.sessions[0]!.mode = mode
const controller = new AbortController()
const cancelled = call(ctx, owner, 'hang', controller.signal)
const queued = call(ctx, owner, 'after cancellation')
setTimeout(() => {
controller.abort(new Error('caller stopped'))
}, 5)
expect((await cancelled).isError).toBe(true)
expect(text(await queued)).toBe('hello from stub')
expect(stub.sessions[0]?.closed).toContain('persistent bash command aborted')
expect(stub.sessions).toHaveLength(2)
},
)
it.each(['init-exit', 'init-timeout'] as const)(
'fails initialization and closes the unusable shell for %s',
async (mode) => {
const { ctx, owner, stub } = await setup({ backendType: 'stub' }, mode)
expect((await call(ctx, owner, 'pwd')).isError).toBe(true)
expect(stub.sessions[0]?.closed).toContain('persistent bash initialization failed')
},
)
it('clears a failed spawn without trying to close an unpublished shell', async () => {
const { ctx, owner, stub } = await setup({ backendType: 'stub' }, 'spawn-error')
expect((await call(ctx, owner, 'pwd')).isError).toBe(true)
expect(stub.sessions).toHaveLength(0)
})
it('resets a cached shell after startSend fails', async () => {
const { ctx, owner, stub } = await setup()
await call(ctx, owner, 'warm up')
stub.sessions[0]!.mode = 'send-error'
expect((await call(ctx, owner, 'fails')).isError).toBe(true)
expect(stub.sessions[0]?.closed).toContain('persistent bash send failed')
expect(text(await call(ctx, owner, 'recovers'))).toBe('hello from stub')
expect(stub.sessions).toHaveLength(2)
})
it('cancels and awaits a pending shell spawn when the plugin is disposed', async () => {
const ctx = new Context()
contexts.push(ctx)
await ctx.plugin(SystemPrompt)
await ctx.plugin(ToolRegistry)
await ctx.plugin(AgentRegistry)
await ctx.plugin(PtyService)
const spawnStarted = Promise.withResolvers<undefined>()
const spawnAborted = Promise.withResolvers<undefined>()
ctx.pty.registerBackend({
type: 'slow',
spawn: spec => new Promise((_resolve, reject) => {
spawnStarted.resolve(undefined)
spec.signal?.addEventListener('abort', () => {
spawnAborted.resolve(undefined)
const reason: unknown = spec.signal?.reason
reject(reason instanceof Error
? reason
: new Error('slow PTY spawn aborted', { cause: reason }))
}, { once: true })
}),
})
const fiber = await ctx.plugin(ToolBashPersistent, { backendType: 'slow' })
const owner = agent(ctx, '/workspace')
const running = call(ctx, owner, 'pwd')
await spawnStarted.promise
await fiber.dispose()
await spawnAborted.promise
expect((await running).isError).toBe(true)
expect(ctx.pty.list(owner)).toEqual([])
})
it('rejects invalid config and invalid calls', async () => {
const { ctx, owner, stub } = await setup()
expect((await call(ctx, undefined, 'pwd')).isError).toBe(true)
expect(text(await call(ctx, owner, ' '))).toContain('command must be a non-empty string')
const controller = new AbortController()
controller.abort(new Error('caller stopped'))
expect((await call(ctx, owner, 'pwd', controller.signal)).isError).toBe(true)
expect(stub.sessions).toHaveLength(0)
expect(() => {
ToolBashPersistent.apply(new Context(), { backendType: '' })
}).toThrow('backendType must be non-empty')
expect(() => {
ToolBashPersistent.apply(new Context(), { timeoutMs: 0 })
}).toThrow('timeoutMs must be a positive safe integer')
expect(() => {
ToolBashPersistent.apply(new Context(), { maxOutputChars: 0 })
}).toThrow('maxOutputChars must be a positive safe integer')
expect(() => {
ToolBashPersistent.apply(new Context(), { description: ' ' })
}).toThrow('description must be non-empty')
})
})

View File

@@ -0,0 +1,16 @@
{
"extends": "../../../tsconfig.base.json",
"compilerOptions": {
"rootDir": "src",
"outDir": "lib/types"
},
"include": ["src"],
"references": [
{ "path": "../../../vendor/cordis" },
{ "path": "../../core/agent" },
{ "path": "../../core/tools" },
{ "path": "../pty" },
{ "path": "../../support/invariants" },
{ "path": "../../util/timeout" }
]
}