fix(headless): dsh run is a direct core front door

This commit is contained in:
Tianyi Cui
2026-08-09 12:13:58 +08:00
parent 772c580ee1
commit 9d5eb37638
159 changed files with 1508 additions and 1042 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/README.md
README.md: 40a888183a9338805d83c86bab9ae16f1c7bc04f
README.zh.md: c0a24d78cb8ff2a2b2ccc2d2a2f4294424be55b0
README.md: 19aed9abafa8e4531d2ee31ff0442003328a9ef6
README.zh.md: 73ed1f500799cb235a8835ffb0317370f9a0a7e6

View File

@@ -2,7 +2,7 @@
English | [中文](README.zh.md)
The session log, system-prompt assembly, tool registry, agent vocabulary, and concrete loop that form the harness's default control spine. These are **product** packages — the stable surface plugins and consumers build against.
The session log, system-prompt assembly, tool registry, agent vocabulary, deployment-default model selection, and concrete loop that form the harness's default control spine. These are **product** packages — the stable surface plugins and consumers build against.
| Package | Role | ctx key |
|---|---|---|
@@ -11,9 +11,10 @@ The session log, system-prompt assembly, tool registry, agent vocabulary, and co
| [`system-prompt/`](system-prompt/README.md) | Prompt and tool-schema assembly registry | `ctx.systemPrompt` |
| [`tools/`](tools/README.md) | Scoped tool registry and execution pipeline | `ctx.tools` |
| [`agent/`](agent/README.md) | Agent interface, registry, and event vocabulary | `ctx.agents` |
| [`agent-default-model/`](agent-default-model/README.md) | Default model selection shared by Agent front doors | `ctx.agentDefaultModel` |
| [`agent-loop/`](agent-loop/README.md) | Default concrete agent driver | `ctx.agentLoop` |
`scope` supplies the shared scoping primitive. `agent` owns the public seam, while `agent-loop` is its default implementation; extension plugins depend on the seam so the driver remains swappable.
`scope` supplies the shared scoping primitive. `agent` owns the public seam, while `agent-loop` is its default implementation; extension plugins depend on the seam so the driver remains swappable. `agent-default-model` owns the deployment selection an Agent front door uses only when a session has no selection of its own.
Runnable compositions belong to [`examples/agent-spine-demo`](../examples/agent-spine-demo/README.md); this group owns only the swappable spine pieces.

View File

@@ -2,7 +2,7 @@
[English](README.md) | 中文
构成 harness 默认控制主干的会话日志、系统提示词组装、工具注册表、agent智能体词汇和具体循环。这些是**产品**包,即插件和消费方构建所依赖的稳定 surface。
构成 harness 默认控制主干的会话日志、系统提示词组装、工具注册表、agent智能体词汇、部署默认模型选择和具体循环。这些是**产品**包,即插件和消费方构建所依赖的稳定 surface。
| 包 | 职责 | ctx key |
|---|---|---|
@@ -11,9 +11,10 @@
| [`system-prompt/`](system-prompt/README.md) | 提示词和工具 schema 组装注册表 | `ctx.systemPrompt` |
| [`tools/`](tools/README.md) | 作用域工具注册表和执行流水线 | `ctx.tools` |
| [`agent/`](agent/README.md) | Agent 接口、注册表和事件词汇 | `ctx.agents` |
| [`agent-default-model/`](agent-default-model/README.md) | 各 Agent 入口共享的默认模型选择 | `ctx.agentDefaultModel` |
| [`agent-loop/`](agent-loop/README.md) | 默认具体 agent 驱动器 | `ctx.agentLoop` |
`scope` 提供共享作用域原语。`agent` 负责公开 seam`agent-loop` 是其默认实现;扩展插件依赖该 seam从而保持驱动器可替换。
`scope` 提供共享作用域原语。`agent` 负责公开 seam`agent-loop` 是其默认实现;扩展插件依赖该 seam从而保持驱动器可替换。`agent-default-model` 负责部署选择Agent 入口仅在会话自身没有选择时使用它。
可运行组合属于 [`examples/agent-spine-demo`](../examples/agent-spine-demo/README.md);该分组只负责可替换的主干组件。

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/core/agent-default-model/README.md
README.md: 02bcc9be3adee2293a20b3ae87ddaf4d52e70deb
README.zh.md: 807b612bd25e49aa318c13c8c8dc7595a6459080

View File

@@ -0,0 +1,25 @@
# @deepseek-ai/dsh-agent-default-model
English | [中文](README.zh.md)
The deployment default used when a front door creates an Agent that has no session-local model selection. `AgentDefaultModelService` provides `ctx.agentDefaultModel`; direct front doors such as `dsh run` and Host-backed front doors such as ApiProxy read the same service instead of owning parallel provider/model defaults.
The plugin config requires `{ provider, model }`. That composition entry is the base of the `agent-default-model` Settings section; a mounted settings provider layers the user's choice over it and changes are visible on the next `currentSelection()` read. `reasoningEffort` belongs to the Settings section but deliberately not to plugin config: a complete saved selection can clear an effort when the next selected model has none, while a composition value would be inherited again.
- `ctx.agentDefaultModel.currentSelection()` returns a detached `{ provider, model, reasoningEffort? }` selection for a newly created Agent.
- `ctx.agentDefaultModel.saveSelection(selection)` saves the complete user selection. Without a settings provider it is a no-op and the composition entry remains current.
The service does not validate catalog membership. A provider route may serve an unadvertised model, and the consumer that actually opens a model request owns availability diagnostics.
## Model Experience
Indirectly, through the provider/model selection supplied to a front door; request assembly and adapters own the model-visible request.
#### KV Cache effect
Changing the default affects only Agents that subsequently resolve from it. An existing session whose request log already names a selection keeps that selection, so this service does not invalidate its established prefix.
## Known Limitations and Deferred Work
- The service owns one process-wide default; per-session selection remains the front door's responsibility.
- Without a settings provider, `saveSelection()` cannot retain a selection for a later Agent.

View File

@@ -0,0 +1,25 @@
# @deepseek-ai/dsh-agent-default-model
[English](README.md) | 中文
该部署默认值供入口在创建尚无会话级模型选择的 Agent 时使用。`AgentDefaultModelService` 提供 `ctx.agentDefaultModel``dsh run` 这类直接入口与 ApiProxy 这类由 Host 支撑的入口读取同一服务,而不是分别持有平行的提供方/模型默认值。
插件配置必须提供 `{ provider, model }`。该组合配置项构成 Settings 中 `agent-default-model` 分节的基础层;挂载的设置提供方在其上叠加用户选择,更改会在下一次调用 `currentSelection()` 时可见。`reasoningEffort` 属于该 Settings 分节但特意不属于插件配置完整保存的选择必须能在下一个选定模型没有推理reasoning强度时清除旧值而组合配置值会再次被继承。
- `ctx.agentDefaultModel.currentSelection()` 返回一份独立的 `{ provider, model, reasoningEffort? }` 选择,供新创建的 Agent 使用。
- `ctx.agentDefaultModel.saveSelection(selection)` 保存完整的用户选择。未挂载设置提供方时,此调用不执行任何操作,组合配置项仍为当前值。
该服务不校验目录成员关系。提供方路由可以服务未在目录中公布的模型;实际发起模型请求的消费方负责可用性诊断。
## 模型体验
通过提供给入口的提供方/模型选择间接影响;模型可见请求由请求组装与适配器负责。
#### KV Cache 影响
更改默认值只影响之后从该默认值解析选择的 Agent。请求日志已经指明选择的现有会话仍沿用该选择因此本服务不会使其已建立的前缀失效。
## 已知限制与暂缓事项
- 该服务只拥有一项进程级默认值;每个会话的选择仍由入口负责。
- 未挂载设置提供方时,`saveSelection()` 无法保留选择供后续 Agent 使用。

View File

@@ -0,0 +1,44 @@
{
"name": "@deepseek-ai/dsh-agent-default-model",
"description": "Default model selection shared by Agent front doors",
"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"
},
"./src/*": "./src/*",
"./package.json": "./package.json"
},
"files": [
"lib/index.js",
"lib/invariant.js",
"lib/types/**/*.d.ts"
],
"license": "BSD-3-Clause",
"dependencies": {
"schemastery": "^3.18.0"
},
"peerDependencies": {
"@deepseek-ai/dsh-agent": "^0.0.1",
"@deepseek-ai/dsh-invariants": "^0.0.1",
"@deepseek-ai/dsh-llm": "^0.0.1",
"@deepseek-ai/dsh-settings": "^0.0.1",
"cordis": "^4.0.0-rc.7"
},
"devDependencies": {
"@deepseek-ai/dsh-agent": "workspace:^",
"@deepseek-ai/dsh-invariants": "workspace:^",
"@deepseek-ai/dsh-llm": "workspace:^",
"@deepseek-ai/dsh-settings": "workspace:^",
"cordis": "^4.0.0-rc.7"
}
}

View File

@@ -0,0 +1,107 @@
/**
* Default model selection for an Agent without a session-specific selection.
*
* @module @deepseek-ai/dsh-agent-default-model
*/
import { Context, Service } from 'cordis'
import z from 'schemastery'
import type { ModelSelection } from '@deepseek-ai/dsh-agent'
import { ReasoningEffortId } from '@deepseek-ai/dsh-llm'
import { installSettingsSection, settingsNamespace } from '@deepseek-ai/dsh-settings'
declare module 'cordis' {
interface Context {
/** Default model selection for Agents created without an explicit model. */
agentDefaultModel: AgentDefaultModelService
}
}
/** Settings namespace carrying the default model selection for future Agents. */
export const AGENT_DEFAULT_MODEL_SETTINGS_NAMESPACE = settingsNamespace('agent-default-model')
/** Stored and composed default model selection. */
export interface AgentDefaultModelSettings {
/** Registered provider route. */
provider: string
/** Provider-owned model id. */
model: string
/** Adapter-owned reasoning effort, or provider/default behavior when absent. */
reasoningEffort?: string
}
/** Schema of the default Agent model settings section. */
export const AGENT_DEFAULT_MODEL_SETTINGS_SCHEMA: z<AgentDefaultModelSettings> = z.object({
provider: z.string().required(),
model: z.string().required(),
reasoningEffort: z.string(),
})
/** Composition entry for the default model selection. */
export interface Config {
/** Registered provider route. */
provider: string
/** Provider-owned model id. */
model: string
}
/** Project stored settings onto the Agent-facing selection type. */
function selection(settings: AgentDefaultModelSettings): ModelSelection {
return {
provider: settings.provider,
model: settings.model,
...settings.reasoningEffort === undefined
? {}
: { reasoningEffort: ReasoningEffortId(settings.reasoningEffort) },
}
}
/**
* Owns the default model selection independently of any Host or transport.
* The composition entry remains usable without a settings provider; when one
* is mounted, its user layer is read live.
*/
export class AgentDefaultModelService extends Service {
static Config: z<Config> = z.object({
provider: z.string().required(),
model: z.string().required(),
})
private source: () => AgentDefaultModelSettings
constructor(ctx: Context, config: Config) {
super(ctx, 'agentDefaultModel')
const entry: AgentDefaultModelSettings = { provider: config.provider, model: config.model }
this.source = () => entry
installSettingsSection(ctx, AGENT_DEFAULT_MODEL_SETTINGS_NAMESPACE, AGENT_DEFAULT_MODEL_SETTINGS_SCHEMA, entry, {
setSource: (current) => { this.source = current },
// Every consumer reads through currentSelection(), so no registration-level fact
// needs rebuilding when the settings document changes.
onChange: () => {},
})
}
/**
* Read the current default model selection.
* @returns a detached provider, model, and optional reasoning selection.
*/
currentSelection(): ModelSelection {
return selection(this.source())
}
/**
* Save the complete default model selection. A deployment without a settings
* provider keeps its composition entry.
* @param next - resolved selection accepted by a front door.
* @returns fulfillment after the optional settings write settles.
*/
async saveSelection(next: ModelSelection): Promise<void> {
await this.ctx.get('settings')?.replace(AGENT_DEFAULT_MODEL_SETTINGS_NAMESPACE, {
provider: next.provider,
model: next.model,
...next.reasoningEffort === undefined ? {} : { reasoningEffort: String(next.reasoningEffort) },
})
}
}
export default AgentDefaultModelService

View File

@@ -0,0 +1,30 @@
/**
* Package-owned invariant companion for the default Agent model selection.
*
* The service owns no independent event relationship: settings registration
* already validates every mutable value before `currentSelection()` can observe it.
* The empty installer keeps that absence explicit in composed invariant sets.
*
* @module @deepseek-ai/dsh-agent-default-model/invariant
*/
import type { Context } from 'cordis'
import type { InvariantInstaller } from '@deepseek-ai/dsh-invariants'
const PACKAGE_NAME = '@deepseek-ai/dsh-agent-default-model'
/** Cordis companion plugin name. */
export const name = 'agent-default-model-invariant'
/** Services required before the companion can register. */
export const inject = ['invariants']
/** No runtime invariant: settings validation owns the only mutable-value relationship. */
const install: InvariantInstaller = () => {}
/**
* Register the intentionally empty invariant contribution.
* @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))

View File

@@ -0,0 +1,98 @@
/** Default Agent model settings layered over a real settings provider. */
import { describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import AgentDefaultModelService, { AGENT_DEFAULT_MODEL_SETTINGS_NAMESPACE } from '../src/index.ts'
import { Settings } from '@deepseek-ai/dsh-settings'
import type { SettingsNamespace } from '@deepseek-ai/dsh-settings'
import { ReasoningEffortId } from '@deepseek-ai/dsh-llm'
/** The smallest real provider: one in-memory document, always writable. */
class MemorySettings extends Settings {
doc: Record<string, unknown> = {}
get writable(): boolean {
return true
}
protected load(): Promise<Record<string, unknown>> {
return Promise.resolve(structuredClone(this.doc))
}
protected persist(ns: SettingsNamespace, section: Record<string, unknown>): Promise<void> {
this.doc = { ...this.doc, [ns]: structuredClone(section) }
return Promise.resolve()
}
}
async function boot(): Promise<{
ctx: Context
settingsFiber: Context['fiber']
defaultModel: AgentDefaultModelService
}> {
const ctx = new Context()
const settingsFiber = ctx.plugin(MemorySettings)
await settingsFiber.await()
await ctx.plugin(AgentDefaultModelService, {
provider: 'deepseek-official',
model: 'deepseek-v4-flash',
})
return { ctx, settingsFiber, defaultModel: ctx.agentDefaultModel }
}
describe('AgentDefaultModelService', () => {
it('resolves the user layer over the composition entry', async () => {
const bench = await boot()
expect(bench.defaultModel.currentSelection()).toEqual({
provider: 'deepseek-official', model: 'deepseek-v4-flash',
})
await bench.defaultModel.saveSelection({
provider: 'acme-gateway', model: 'acme-large', reasoningEffort: ReasoningEffortId('high'),
})
expect(bench.defaultModel.currentSelection()).toEqual({
provider: 'acme-gateway', model: 'acme-large', reasoningEffort: 'high',
})
await bench.ctx.fiber.dispose()
})
it('clears a stored effort when the saved selection has none', async () => {
const bench = await boot()
await bench.defaultModel.saveSelection({
provider: 'acme-gateway', model: 'acme-large', reasoningEffort: ReasoningEffortId('high'),
})
await bench.defaultModel.saveSelection({ provider: 'acme-gateway', model: 'acme-plain' })
expect(bench.defaultModel.currentSelection()).toEqual({ provider: 'acme-gateway', model: 'acme-plain' })
await bench.ctx.fiber.dispose()
})
it('layers a hand-written partial section over the entry', async () => {
const bench = await boot()
await bench.settingsFiber.ctx.settings.replace(AGENT_DEFAULT_MODEL_SETTINGS_NAMESPACE, {
model: 'deepseek-reasoner',
})
expect(bench.defaultModel.currentSelection()).toEqual({
provider: 'deepseek-official', model: 'deepseek-reasoner',
})
await bench.ctx.fiber.dispose()
})
it('falls back to the composition entry when the settings provider detaches', async () => {
const bench = await boot()
await bench.defaultModel.saveSelection({ provider: 'acme-gateway', model: 'acme-large' })
expect(bench.defaultModel.currentSelection().provider).toBe('acme-gateway')
await bench.settingsFiber.dispose()
expect(bench.defaultModel.currentSelection()).toEqual({
provider: 'deepseek-official', model: 'deepseek-v4-flash',
})
await bench.ctx.fiber.dispose()
})
it('keeps the composition entry when no settings provider is mounted', async () => {
const ctx = new Context()
await ctx.plugin(AgentDefaultModelService, { provider: 'p', model: 'm' })
await ctx.agentDefaultModel.saveSelection({ provider: 'other', model: 'other' })
expect(ctx.agentDefaultModel.currentSelection()).toEqual({ provider: 'p', model: 'm' })
await ctx.fiber.dispose()
})
})

View File

@@ -0,0 +1,30 @@
{
"extends": "../../../tsconfig.base.json",
"compilerOptions": {
"rootDir": "src",
"outDir": "lib/types"
},
"include": [
"src"
],
"references": [
{
"path": "../../../vendor/cordis"
},
{
"path": "../../../vendor/schemastery"
},
{
"path": "../../core/agent"
},
{
"path": "../../llm/llm"
},
{
"path": "../../settings/settings"
},
{
"path": "../../support/invariants"
}
]
}

View File

@@ -0,0 +1,25 @@
import { defineConfig } from 'tsdown'
/** Build the package root and invariant companion as independent bundles. */
export default defineConfig([
{
entry: ['lib/types/index.js'],
outDir: 'lib',
format: ['esm'],
platform: 'node',
target: 'es2024',
fixedExtension: false,
dts: false,
clean: false,
},
{
entry: ['lib/types/invariant.js'],
outDir: 'lib',
format: ['esm'],
platform: 'node',
target: 'es2024',
fixedExtension: false,
dts: false,
clean: false,
},
])

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/agent/README.md
README.md: 3a3bdf6a4b3bdc5bfef250495e84b7d90b003822
README.zh.md: e421070a6eec2e6e7e1fc7b45f0a5e29040bf712
README.md: ba0fa456593e598b6532c10ef2390bbd55f3a486
README.zh.md: 581da5c8621e5653f57edce38ac9816da7a40d40

View File

@@ -12,7 +12,7 @@ Tracks live agents and carries the initiating Agent through asynchronous driver
### Public API
The scoped-registration surface: `Agent.ctx` is the agent's scope context (`dsh-scope`, key = the agent) — register tools/sections/variables/listeners through it for that agent alone, all unwound on disposal. `agentEvents(ctx, agent)` is the fused dispatcher for ordinary agent-subject operations (carrier + injected subject in one move); its notification mode invokes every listener and contains both synchronous throws and returned-promise rejections. The registry lifecycle pair reuses one stable routing carrier. `assembleContextFor(agent)` builds the per-agent assembly context (`agent` + `scope` together). `installAgentLlmTarget(agentCtx, target)` snapshots a mutable provider/model/reasoning-effort selection during prompt assembly, applies the route to prompt variables, and applies the complete target to request routing for one step; an absent selected effort clears an inherited effort so the target uses adapter/provider defaults. `CreateAgentOptions.setup(agentCtx)` and `ResumeAgentOptions.setup(agentCtx)` compose a fresh or resumed agent's scoped world while both objects remain unpublished. Setup is trusted, composition-only same-process code: drive the agent only after creation resolves.
The scoped-registration surface: `Agent.ctx` is the agent's scope context (`dsh-scope`, key = the agent) — register tools/sections/variables/listeners through it for that agent alone, all unwound on disposal. `agentEvents(ctx, agent)` is the fused dispatcher for ordinary agent-subject operations (carrier + injected subject in one move); its notification mode invokes every listener and contains both synchronous throws and returned-promise rejections. The registry lifecycle pair reuses one stable routing carrier. `assembleContextFor(agent)` builds the per-agent assembly context (`agent` + `scope` together). `installModelSelection(agentCtx, selection)` snapshots a mutable provider/model/reasoning-effort selection during prompt assembly, applies its provider and model to prompt variables, and applies the complete selection to request routing for one step; an absent selected effort clears an inherited effort so adapter/provider defaults apply. `CreateAgentOptions.setup(agentCtx)` and `ResumeAgentOptions.setup(agentCtx)` compose a fresh or resumed agent's scoped world while both objects remain unpublished. Setup is trusted, composition-only same-process code: drive the agent only after creation resolves.
`AgentOptions` supplies the initial provider/model route and an optional positive `maxTokens` output cap. The concrete loop resolves any exact-model adapter default, records the effective cap in the request header, and applies it to each conversation-model request; an explicit Agent option wins, while omission leaves the adapter or provider route default in control.

View File

@@ -12,7 +12,7 @@ Agent 接口、注册表、进程本地发起方作用域,以及 `agent/*` 事
### 公开 API
带作用域的注册接口:`Agent.ctx` 是 agent 的作用域上下文(`dsh-scope`,键 = 该 agent。通过它注册工具变量监听器只对该 agent 生效,并在 dispose资源释放时全部撤销。`agentEvents(ctx, agent)` 是普通 agent 主体操作的融合分发器(一次完成载体 + 注入主体);其通知 mode 会调用每个监听器,并同时收容同步抛出和返回 Promise 的拒绝。注册表生命周期对复用一个稳定路由载体。`assembleContextFor(agent)` 构建按 agent 的组装上下文(同时包含 `agent` + `scope`)。`installAgentLlmTarget(agentCtx, target)` 在提示词组装期间快照可变的提供方模型推理reasoning强度选择路由应用到提示词变量,并将完整目标应用到一个步骤的请求路由;如果没有选定推理强度,则会清除继承的推理强度,使该目标使用适配器/提供方默认值。`CreateAgentOptions.setup(agentCtx)``ResumeAgentOptions.setup(agentCtx)` 在新建或恢复的 agent 尚未发布时组合其带作用域的世界。Setup 是受信任、仅用于组合的同进程代码:只有创建完成后才能驱动 agent。
带作用域的注册接口:`Agent.ctx` 是 agent 的作用域上下文(`dsh-scope`,键 = 该 agent。通过它注册工具变量监听器只对该 agent 生效,并在 dispose资源释放时全部撤销。`agentEvents(ctx, agent)` 是普通 agent 主体操作的融合分发器(一次完成载体 + 注入主体);其通知 mode 会调用每个监听器,并同时收容同步抛出和返回 Promise 的拒绝。注册表生命周期对复用一个稳定路由载体。`assembleContextFor(agent)` 构建按 agent 的组装上下文(同时包含 `agent` + `scope`)。`installModelSelection(agentCtx, selection)` 在提示词组装期间快照可变的提供方模型推理reasoning强度选择其中的提供方和模型应用到提示词变量,并将完整选择应用到一个步骤的请求路由;如果没有选定推理强度,则会清除继承的推理强度,使适配器/提供方默认值生效`CreateAgentOptions.setup(agentCtx)``ResumeAgentOptions.setup(agentCtx)` 在新建或恢复的 agent 尚未发布时组合其带作用域的世界。Setup 是受信任、仅用于组合的同进程代码:只有创建完成后才能驱动 agent。
`AgentOptions` 提供初始的提供方/模型路由,以及可选的正数 `maxTokens` 输出上限。具体循环会解析确切模型的适配器默认值,把生效上限记录到请求 header并应用到每次对话模型请求显式 Agent 选项优先,省略时由适配器或提供方路由默认值控制。

View File

@@ -17,7 +17,7 @@ import type { Agent, AgentOptions } from './types.ts'
export * from './types.ts'
export * from './inbox.ts'
export * from './llm-target.ts'
export * from './model-selection.ts'
export { agentCarrier, agentEvents, assembleContextFor, emitAgentEvent } from './dispatch.ts'
export type { AgentEventDispatch, AgentSubjectEvent } from './dispatch.ts'

View File

@@ -1,13 +1,13 @@
/**
* Agent-scoped LLM target snapshot shared by interactive front doors.
* @module @deepseek-ai/dsh-agent/llm-target
* Agent-scoped model selection shared by interactive front doors.
* @module @deepseek-ai/dsh-agent/model-selection
*/
import type { Context } from 'cordis'
import type { LlmCallConfig, ReasoningEffortId } from '@deepseek-ai/dsh-llm'
/** Complete provider/model route and optional reasoning effort selected for one live agent. */
export interface AgentLlmTarget {
/** Complete provider, model, and optional reasoning effort selected for one live Agent. */
export interface ModelSelection {
/** Registered provider route. */
provider: string
/** Provider-owned model id. */
@@ -16,31 +16,31 @@ export interface AgentLlmTarget {
reasoningEffort?: ReasoningEffortId
}
/** Mutable selection plus the target captured for the current step. */
export interface AgentLlmTargetRef {
/** Target selected for the next step that enters prompt assembly. */
current: AgentLlmTarget | undefined
/** Target captured when the current step entered prompt assembly. */
assembled: AgentLlmTarget | undefined
/** Mutable model selection plus the value captured for the current step. */
export interface ModelSelectionRef {
/** Model selected for the next step that enters prompt assembly. */
current: ModelSelection | undefined
/** Selection captured when the current step entered prompt assembly. */
assembled: ModelSelection | undefined
}
/**
* Couple one mutable target to agent-scoped prompt assembly and request routing.
* Prompt assembly snapshots the selected target before delegating, then applies
* its route to prompt variables and its route/effort to request config so a
* Couple one mutable selection to Agent-scoped prompt assembly and request routing.
* Prompt assembly snapshots the selected model before delegating, then applies
* its provider/model pair and effort to request config so a
* concurrent switch takes effect on a later step instead of splitting the two
* surfaces. An absent selected effort clears any inherited effort so a model
* switch can restore that target's provider/default behavior.
* surfaces. An absent selected effort clears any inherited effort, restoring
* the selected model's provider/default behavior.
*
* @param agentCtx - The target agent's scoped context.
* @param target - Mutable selection owned by the calling front door.
* @param agentCtx - The selected Agent's scoped context.
* @param selection - Mutable selection owned by the calling front door.
* @returns Disposer for both scoped waterfall listeners.
*/
export function installAgentLlmTarget(agentCtx: Context, target: AgentLlmTargetRef): () => void {
export function installModelSelection(agentCtx: Context, selection: ModelSelectionRef): () => void {
const disposeAssembly = agentCtx.on('system-prompt/assemble', async (_assembly, _context, next) => {
const selected = target.current
const selected = selection.current
const assembled = await next()
target.assembled = selected
selection.assembled = selected
if (selected === undefined) return assembled
return {
...assembled,
@@ -55,7 +55,7 @@ export function installAgentLlmTarget(agentCtx: Context, target: AgentLlmTargetR
'agent/request',
async (_payload, next): Promise<LlmCallConfig> => {
const resolved = await next()
const selected = target.assembled
const selected = selection.assembled
if (selected === undefined) return resolved
const { reasoningEffort: _inheritedEffort, ...withoutInheritedEffort } = resolved
return {

View File

@@ -3,18 +3,18 @@ import { Context } from 'cordis'
import SystemPrompt from '@deepseek-ai/dsh-system-prompt'
import {
agentEvents,
installAgentLlmTarget,
installModelSelection,
type Agent,
type AgentLlmTargetRef,
type ModelSelectionRef,
} from '../src/index.ts'
import { ReasoningEffortId, type LlmCallConfig } from '@deepseek-ai/dsh-llm'
describe('installAgentLlmTarget()', () => {
describe('installModelSelection()', () => {
it('snapshots prompt variables and request routing together, then disposes both listeners', async () => {
const ctx = new Context()
await ctx.plugin(SystemPrompt)
const target: AgentLlmTargetRef = { current: undefined, assembled: undefined }
const dispose = installAgentLlmTarget(ctx, target)
const selection: ModelSelectionRef = { current: undefined, assembled: undefined }
const dispose = installModelSelection(ctx, selection)
const agent = {} as Agent
const seed: LlmCallConfig = { provider: 'seed', model: 'seed', temperature: 0.2 }
const signal = new AbortController().signal
@@ -24,13 +24,13 @@ describe('installAgentLlmTarget()', () => {
'agent/request', { turn: 1, step: 0, signal }, () => Promise.resolve(seed),
)).resolves.toBe(seed)
target.current = {
selection.current = {
provider: 'alpha',
model: 'a1',
reasoningEffort: ReasoningEffortId('high'),
}
expect((await ctx.systemPrompt.assemble()).variables).toMatchObject({ provider: 'alpha', model: 'a1' })
target.current = { provider: 'beta', model: 'b1' }
selection.current = { provider: 'beta', model: 'b1' }
await expect(agentEvents(ctx, agent).waterfall(
'agent/request', { turn: 1, step: 0, signal }, () => Promise.resolve(seed),
)).resolves.toEqual({