Merge branch 'codex/tool-json-schema-dsl' into codex/canonical-tool-output

# Conflicts:
#	docs/event-producer-consumer.md
#	packages/context/time-context/tests/time-context.spec.ts
#	packages/context/workspace-context/tests/workspace-context.spec.ts

examples/acp-agent/cordis.yml

@@ -9,6 +9,11 @@
   config:
     apiKey: !!js process.env.DEEPSEEK_API_KEY
     baseURL: !!js process.env.DEEPSEEK_BASE_URL
+    models:
+      - id: deepseek-v4-flash
+        contextWindow: 256000
+      - id: deepseek-v4-pro
+        contextWindow: 256000
 
 # The default composition confines bash AND the filesystem tools to the
 # workspace and asks before a wider retry. Snapshot runs select
@@ -58,20 +63,17 @@
 
       Verify your work by running the code or tests. Keep answers brief and factual.
 
-# Replay-aware request pressure with one service-wide context window.
+# Replay-aware request pressure; the routed adapter supplies model capacity.
 - id: token-meter
   name: '@deepseek-ai/dsh-token-meter'
-  config:
-    # FIXME: Resolve compaction config per model; this capacity assumes a 256k context window.
-    contextWindow: 256000
 
 # Summarize an older range after measured pressure or a canonical provider overflow.
-# Service-wide policy provides pressure, retention, and one overflow-retry default.
+# Ratios scale against the routed model's context window.
 - id: compact-basic
   name: '@deepseek-ai/dsh-compact-basic'
   config:
     thresholdRatio: 0.8
-    retainTokens: 20480
+    retainRatio: 0.08
     maxTokens: 8192
     compactionRetries: 1
 

examples/headless-agent/composition.md

@@ -21,6 +21,8 @@ flowchart LR
   bundle_agent_core --> spine_sessions["ctx.sessions"]
   bundle_agent_core --> spine_tools["ctx.tools + tool-bash"]
   bundle_agent_core --> spine_loop["ctx.agents + ctx.agentLoop"]
+  plugin_headless_token_meter["token-meter<br/>@deepseek-ai/dsh-token-meter"]
+  cfg --> plugin_headless_token_meter
   plugin_headless_compact_basic["compact-basic<br/>@deepseek-ai/dsh-compact-basic"]
   cfg --> plugin_headless_compact_basic
   plugin_headless_subagent["subagent<br/>@deepseek-ai/dsh-subagent"]
@@ -54,6 +56,7 @@ flowchart LR
 | `llm-deepseek` | `@deepseek-ai/dsh-llm-deepseek` |
 | `bash` | `@deepseek-ai/dsh-bash-local` |
 | `cli-agent` | `@deepseek-ai/dsh-cli-demo` |
+| `token-meter` | `@deepseek-ai/dsh-token-meter` |
 | `compact-basic` | `@deepseek-ai/dsh-compact-basic` |
 | `subagent` | `@deepseek-ai/dsh-subagent` |
 | `subagent-spawn` | `@deepseek-ai/dsh-subagent-spawn` |

examples/headless-agent/cordis.yml

@@ -11,7 +11,9 @@
     baseURL: !!js process.env.DEEPSEEK_BASE_URL
     models:
       - id: deepseek-v4-pro
+        contextWindow: 128000
       - id: deepseek-v4-flash
+        contextWindow: 128000
 
 - id: bash
   name: '@deepseek-ai/dsh-bash-local'
@@ -35,13 +37,14 @@
       factual.
 
 # Summarize an older range when derived history approaches the context window.
+- id: token-meter
+  name: '@deepseek-ai/dsh-token-meter'
+
 - id: compact-basic
   name: '@deepseek-ai/dsh-compact-basic'
   config:
-    contextWindow: 128000
     thresholdRatio: 0.8
-    retainTokens: 20480
-    summarizationModel: ''
+    retainRatio: 0.16
     maxTokens: 8192
     compactionRetries: 1
 

examples/headless-agent/tests/compaction.e2e.ts

@@ -33,9 +33,7 @@ describe.skipIf(!process.env.DEEPSEEK_API_KEY)('compaction: a long session compa
     // Reasoning tokens require a larger generation cap than the retained checkpoint.
     ctx = await codingHarness(workdir, {
       persona: SYSTEM_PROMPT,
-      tokenMeter: {
-        contextWindow: 2000,
-      },
+      modelContextWindow: 2000,
       compact: {
         thresholdRatio: 0.5,
         retainTokens: 400,

examples/headless-agent/tests/harness.ts

@@ -8,7 +8,6 @@ import * as ToolBash from '@deepseek-ai/dsh-tool-bash'
 import * as ToolTodo from '@deepseek-ai/dsh-tool-todo'
 import * as LlmDeepSeek from '@deepseek-ai/dsh-llm-deepseek'
 import TokenMeterService from '@deepseek-ai/dsh-token-meter'
-import type { TokenMeterConfig } from '@deepseek-ai/dsh-token-meter'
 import ToolResultPruneService from '@deepseek-ai/dsh-compact-tool-result-prune'
 import SessionPersistenceJsonl from '@deepseek-ai/dsh-session-persistence-jsonl'
 import { BasicCompactService } from '@deepseek-ai/dsh-compact-basic'
@@ -46,8 +45,8 @@ export interface CodingHarnessOptions {
    * compaction plugin (the default suites run without it).
    */
   compact?: BasicCompactConfig
-  /** Optional token-meter capacity loaded before compact-basic. */
-  tokenMeter?: TokenMeterConfig
+  /** Test-only context capacity advertised for `deepseek-v4-flash`. */
+  modelContextWindow?: number
 }
 
 export async function codingHarness(workdir: string, options: CodingHarnessOptions = {}): Promise<Context> {
@@ -56,14 +55,15 @@ export async function codingHarness(workdir: string, options: CodingHarnessOptio
     systemPrompt: { persona: options.persona ?? '' },
   })
   await ctx.plugin(AgentLoop, { agents: [] })
-  await ctx.plugin(LlmDeepSeek)
+  await ctx.plugin(LlmDeepSeek, options.modelContextWindow === undefined ? {} : {
+    models: [{ id: 'deepseek-v4-flash', contextWindow: options.modelContextWindow }],
+  })
   await ctx.plugin(LocalBashExecutor, { cwd: workdir, timeoutMs: 30_000 })
   await ctx.plugin(ToolBash)
   await ctx.plugin(ToolTodo)
-  // Compaction is opt-in: only the compaction e2e loads the reusable meter and
-  // backend, with a lower context window so a short real session crosses the threshold.
+  // Compaction is opt-in: only the compaction e2e loads the reusable meter and backend.
   if (options.compact !== undefined) {
-    await ctx.plugin(TokenMeterService, options.tokenMeter)
+    await ctx.plugin(TokenMeterService)
     await ctx.plugin(ToolResultPruneService)
     await ctx.plugin(BasicCompactService, options.compact)
   }

examples/jsonrpc-agent/cordis.yml

@@ -63,12 +63,13 @@
 - id: tool-fs
   name: '@deepseek-ai/dsh-tool-fs'
 
+- id: token-meter
+  name: '@deepseek-ai/dsh-token-meter'
+
 - id: compact-basic
   name: '@deepseek-ai/dsh-compact-basic'
   config:
-    contextWindow: 128000
     thresholdRatio: 0.8
-    retainTokens: 20480
-    summarizationModel: ''
+    retainRatio: 0.16
     maxTokens: 8192
     compactionRetries: 1

examples/tui-agent/tests/fixtures/tui-scripted-llm.ts

@@ -1,5 +1,5 @@
 import type { Context } from 'cordis'
-import type { GenerateOptions, LlmModelInfo, StreamChunk } from '@deepseek-ai/dsh-llm'
+import type { GenerateOptions, LlmModelContext, LlmModelInfo, StreamChunk } from '@deepseek-ai/dsh-llm'
 import { CallId, LlmAdapter } from '@deepseek-ai/dsh-llm'
 
 const CONTROL_PROBE = '\u001b]2;MODEL_CONTROLLED\u0007\u001b[999CMODEL_CURSOR\u009b31mMODEL_C1'
@@ -25,6 +25,10 @@ class ScriptedTuiAdapter extends LlmAdapter {
     ])
   }
 
+  override resolveModelContext(_provider: string, _model: string): Promise<LlmModelContext> {
+    return Promise.resolve({ contextWindow: 128_000 })
+  }
+
   override async * stream(options: GenerateOptions): AsyncIterable<StreamChunk> {
     if (options.model !== 'tui-scripted-model-pro' || !options.system?.includes('tui-scripted-model-pro')) {
       throw new Error('the scripted TUI request did not apply the selected model to routing and prompt variables')

examples/tui-agent/tests/tui.snapshot.ts

@@ -33,7 +33,7 @@ import { HeadlessTerminal } from '../../../packages/ui/tui/tests/headless-termin
 const SNAPSHOTS_DIR = join(dirname(fileURLToPath(import.meta.url)), 'snapshots')
 // Keep pre-normalization layout widths identical across macOS and Linux.
 const SNAPSHOT_TMP_ROOT = process.platform === 'win32' ? tmpdir() : '/tmp'
-const PROVIDERS = [{ id: 'deepseek', models: [{ id: 'deepseek-v4-flash' }] }]
+const PROVIDERS = [{ id: 'deepseek', models: [{ id: 'deepseek-v4-flash', contextWindow: 128_000 }] }]
 const UUID_RE = /[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}/gi
 
 type SnapshotMode = 'replay' | 'record' | 'refresh'