dsh-model-router 0.2.0 → 0.4.0

package/README.md

@@ -46,6 +46,21 @@ Everything else is a planner. That logic lives in `src/policy.ts` as a plain fun
 
 The router always stamps `provider` + `model`. `reasoningEffort` and `maxTokens` are *optional per role*: set them in the config and they're enforced for that role; leave them out and those fields inherit from your session's selection. So picking "max effort" in the UI but not pinning `reasoningEffort` in the config still gives you max-effort thinking — it just happens on the routed model.
 
+## Turning the router off
+
+Routing is on by default. Two ways to switch it off:
+
+- **GUI (Settings → Plugins → dsh-model-router):** the plugin registers a live
+  settings section; flip `enabled` off. It applies immediately (no restart),
+  persists in `settings.yaml` under `model-router:`, and unregisters the prompt
+  section and the skill too. Flip it back on and everything returns.
+- **Patch row:** set `enabled: false` in the profile's `cordis.patch.yml` row
+  (takes effect on the next boot). `disabled: true` still skips the row entirely.
+
+With the router off, requests use the session's selected model (your
+`agent-default-model` setting or the base default) — the router is simply not
+rewriting them.
+
 ## Tuning
 
 All configuration lives on the plugin row. Patch it in the profile's `cordis.patch.yml`:
@@ -62,7 +77,10 @@ All configuration lives on the plugin row. Patch it in the profile's `cordis.pat
         executor:           # subagent route
           provider: deepseek-official
           model: deepseek-v4-flash
-          reasoningEffort: low
+          reasoningEffort: high
+          escalateOnError: true   # after a failed step…
+          escalateTo: max         #   …bump effort for the next request
+          recoverySteps: 2        #   …wearing off after N clean steps
         mode: strict        # strict | plan (see below)
         promptSection: true # register the always-on routing section
         skill: true         # register the pro-flash-routing skill
@@ -70,6 +88,8 @@ All configuration lives on the plugin row. Patch it in the profile's `cordis.pat
 
 `mode` controls how the root agent is treated: `strict` keeps it on the planner route always; `plan` sends the root to the executor route unless plan mode is active, reserving pro for real planning.
 
+**Error-driven escalation** (`escalateOnError`): when a route's agent hits a failed tool step, the *next* request bumps to `escalateTo` and wears off after `recoverySteps` clean steps. It's deterministic and stateless — the router folds the session log per request, so only prior steps are considered (a failure can't escalate the very request that caused it). It's a per-route knob: enable it on the executor to make flash think harder after a flubbed execution step, without touching the baseline.
+
 The defaults are exactly the table at the top of this page. To switch the router off for a session, disable the row (`disabled: true`) or remove the plugin — `dsh plugin --profile web remove dsh-model-router`.
 
 ## Reduce pro token usage

package/cordis.patch.yml

@@ -19,12 +19,16 @@
         executor:
           provider: deepseek-official
           model: deepseek-v4-flash
-          # reasoningEffort: low
+          # reasoningEffort: high
           # maxTokens: 16384
+          # escalateOnError: true   # v1: after a failed step, escalate effort
+          # escalateTo: max         #   to this level for the next request…
+          # recoverySteps: 2        #   …wearing off after N clean steps
         # strict: root is always the planner (pro).
         # plan:   root is pro only while plan mode is active; otherwise it
         #         falls back to the executor route to reserve pro for planning.
         mode: strict
+        # enabled: false  # off switch — GUI: Settings → Plugins → model-router (live, no restart); patch row needs a reboot
         # Publish the always-on routing convention prompt section.
         promptSection: true
         # Register the `pro-flash-routing` skill in the session catalog.

package/lib/index.d.ts

@@ -1,3 +1,4 @@
+import * as _deepseek_ai_dsh_settings from '@deepseek-ai/dsh-settings';
 import { Service, Context } from '@deepseek-ai/cordis';
 import z from '@deepseek-ai/schemastery';
 
@@ -26,6 +27,17 @@ interface ModelRoute {
     reasoningEffort?: ReasoningEffort;
     /** Optional output-token cap for the role; omitted means inherit. */
     maxTokens?: number;
+    /**
+     * Error-driven escalation (v1): when true, a failed execution step bumps the
+     * next request's effort to `escalateTo`, wearing off after `recoverySteps`
+     * clean steps. Deterministic, stateless — the session log is folded per
+     * request, so only *prior* steps are ever considered.
+     */
+    escalateOnError?: boolean;
+    /** Effort used for the request after a failed step. */
+    escalateTo?: ReasoningEffort;
+    /** Clean steps before escalation wears off. Defaults to 2. */
+    recoverySteps?: number;
 }
 /** The two roles the router distinguishes. */
 type AgentRole = "planner" | "executor";
@@ -34,6 +46,13 @@ interface RouterConfig {
     planner: ModelRoute;
     executor: ModelRoute;
     mode: RoutingMode;
+    /**
+     * Live off-switch. Defaults to true; settable from Settings → Plugins →
+     * model-router (applies immediately) or from the patch row (next boot).
+     * When false, the router stops rewriting requests and unregisters the
+     * prompt section and the skill.
+     */
+    enabled: boolean;
     promptSection: boolean;
     skill: boolean;
 }
@@ -58,27 +77,33 @@ declare function roleFor(agent: unknown): AgentRole;
  * @returns the model route to stamp, or `undefined` to leave the request alone.
  */
 declare function routeFor(agent: unknown, config: RouterConfig, planModeActive?: boolean): ModelRoute | undefined;
-
 /**
- * dsh-model-router: role-based model routing for the DeepSeek Harness.
+ * Whether any of the last `recoverySteps` completed steps carried a failed
+ * tool result. A failure is a `tool/result` event whose data carries an
+ * `error` field (the harness records tool failures there).
  *
- * The planner (the session's root agent) runs on `deepseek-v4-pro`; delegated
- * executor subagents run on `deepseek-v4-flash`. Enforcement is a per-agent
- * `agent/request` rewrite registered when the agent is created, so it applies
- * in every mode (web / headless / tui) and every agent preset, including
- * subagents the delegation tools create.
+ * Steps are deduplicated by `turn:step`, and only *completed* steps count —
+ * events are scanned from the tail, so the current in-flight request is never
+ * considered.
  *
- * Each role route may also pin `reasoningEffort` and `maxTokens`; when set,
- * they override the session's selection for that role. A `mode` switch lets a
- * deployment reserve the planner route for actual planning.
+ * @param events - the agent's session event log (or `undefined`).
+ * @param recoverySteps - how many completed steps back to scan.
+ * @returns true when a failed step is within the window.
+ */
+declare function recentStepsHadError(events: readonly unknown[] | undefined, recoverySteps?: number): boolean;
+/**
+ * Resolve the reasoning effort to stamp for one request.
  *
- * The plugin also publishes:
- *  - a system-prompt section stating the planner/executor convention, and
- *  - the `pro-flash-routing` skill teaching the agent to plan itself and
- *    delegate code execution to flash subagents.
+ * Baseline is the route's `reasoningEffort`; when `escalateOnError` is enabled
+ * and a recent step failed, the effort bumps to `escalateTo` (falling back to
+ * the baseline when `escalateTo` is unset). Returns `undefined` to leave the
+ * request's effort alone (inherit the session selection).
  *
- * @module dsh-model-router
+ * @param route - the resolved route for the agent.
+ * @param events - the agent's session event log.
+ * @returns the effort to stamp, or `undefined` to inherit.
  */
+declare function effortFor(route: ModelRoute, events: readonly unknown[] | undefined): ReasoningEffort | undefined;
 
 /** Plugin row id; the bundle patch inserts it under this id. */
 declare const name = "model-router";
@@ -89,24 +114,37 @@ declare const Config: z<Schemastery.ObjectS<{
         model: z<string, string>;
         reasoningEffort: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
         maxTokens: z<number, number>;
+        escalateOnError: z<boolean, boolean>;
+        escalateTo: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
+        recoverySteps: z<number, number>;
     }>, Schemastery.ObjectT<{
         provider: z<string, string>;
         model: z<string, string>;
         reasoningEffort: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
         maxTokens: z<number, number>;
+        escalateOnError: z<boolean, boolean>;
+        escalateTo: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
+        recoverySteps: z<number, number>;
     }>>;
     executor: z<Schemastery.ObjectS<{
         provider: z<string, string>;
         model: z<string, string>;
         reasoningEffort: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
         maxTokens: z<number, number>;
+        escalateOnError: z<boolean, boolean>;
+        escalateTo: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
+        recoverySteps: z<number, number>;
     }>, Schemastery.ObjectT<{
         provider: z<string, string>;
         model: z<string, string>;
         reasoningEffort: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
         maxTokens: z<number, number>;
+        escalateOnError: z<boolean, boolean>;
+        escalateTo: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
+        recoverySteps: z<number, number>;
     }>>;
     mode: z<"strict" | "plan", "strict" | "plan">;
+    enabled: z<boolean, boolean>;
     promptSection: z<boolean, boolean>;
     skill: z<boolean, boolean>;
 }>, Schemastery.ObjectT<{
@@ -115,40 +153,112 @@ declare const Config: z<Schemastery.ObjectS<{
         model: z<string, string>;
         reasoningEffort: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
         maxTokens: z<number, number>;
+        escalateOnError: z<boolean, boolean>;
+        escalateTo: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
+        recoverySteps: z<number, number>;
     }>, Schemastery.ObjectT<{
         provider: z<string, string>;
         model: z<string, string>;
         reasoningEffort: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
         maxTokens: z<number, number>;
+        escalateOnError: z<boolean, boolean>;
+        escalateTo: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
+        recoverySteps: z<number, number>;
     }>>;
     executor: z<Schemastery.ObjectS<{
         provider: z<string, string>;
         model: z<string, string>;
         reasoningEffort: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
         maxTokens: z<number, number>;
+        escalateOnError: z<boolean, boolean>;
+        escalateTo: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
+        recoverySteps: z<number, number>;
     }>, Schemastery.ObjectT<{
         provider: z<string, string>;
         model: z<string, string>;
         reasoningEffort: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
         maxTokens: z<number, number>;
+        escalateOnError: z<boolean, boolean>;
+        escalateTo: z<"off" | "low" | "high" | "max", "off" | "low" | "high" | "max">;
+        recoverySteps: z<number, number>;
     }>>;
     mode: z<"strict" | "plan", "strict" | "plan">;
+    enabled: z<boolean, boolean>;
     promptSection: z<boolean, boolean>;
     skill: z<boolean, boolean>;
 }>>;
+/** Settings namespace the live on/off toggle lives under (settings.yaml). */
+declare const SETTINGS_NS: _deepseek_ai_dsh_settings.SettingsNamespace;
 declare const SKILL_NAME = "pro-flash-routing";
 declare const SKILL_DESCRIPTION = "Route planning and code execution across models: plan on the pro planner agent, delegate implementation to flash executor subagents.";
 declare const SKILL_WHEN_TO_USE = "Use when a task combines planning and implementation: before writing code, after a plan is approved, when delegating execution work, or when the user asks about the pro/flash routing convention.";
 declare const SKILL_CONTENT = "# Pro planner / Flash executor routing\n\nThis session routes models by role:\n\n- **Planner (this agent)** \u2014 `deepseek-v4-pro`. Planning, design decisions, reviewing delegated output, and user-facing synthesis happen here.\n- **Executors (every subagent)** \u2014 `deepseek-v4-flash`. Implementation work happens there: writing code, running commands, builds, and tests. The harness forces the model automatically; you do not select it.\n\n## Working rhythm\n\n1. **Plan here.** Explore, decide the approach, and (when plan mode is on) submit the plan with `exit_plan_mode`. The plan stays on this agent.\n2. **Delegate the execution.** Once a plan is approved, hand each self-contained chunk of implementation to a subagent with a complete prompt: exact files to touch, the change to make, and how to verify. Subagents are automatically routed to `deepseek-v4-flash`, so keep them execution-focused: give them the decision, not the decision to make.\n3. **Review here.** Read the subagent's result on this agent, verify it yourself (tests, diffs, logs), and iterate with follow-up messages to the same subagent when available.\n4. **Report here.** Summaries, plans, and answers to the user come from this agent.\n\n## Keep this agent's context lean\n\nInput tokens are the expensive part of the planner. Don't re-read large files or full transcripts on this agent \u2014 trust the subagent's final report. Prefer targeted reads (offset/limit) over whole files. When the context grows, compact rather than re-sending everything.\n\n## Delegation guidelines\n\n- Start independent delegations together in one assistant message and continue useful work while they run (background mode by default).\n- Prefer `subagent` for self-contained work and `workflow` when many independent pieces need fan-out; their workers run on flash as well.\n- Do not delegate design: subagents execute decisions already made.\n- If a subagent's task grows into design work, pull it back to this agent and re-delegate the narrowed execution.\n\n## Verification\n\n- Executor output was produced by `deepseek-v4-flash`; planner output by `deepseek-v4-pro`. If you need to confirm, check the session log's model metadata.\n- If routing ever looks wrong, the `model-router` plugin row in the profile composition is the single place that owns it.";
 /** The plugin row id the bundle patch must insert. */
 declare const ROW_ID = "model-router";
+/** Minimal structural view of the live agent object the router reads. */
+interface AgentLike {
+    ctx: AgentScopedContext;
+    options?: {
+        subagentDepth?: number;
+    };
+    session?: {
+        header?: {
+            origin?: string;
+        };
+        events?: unknown[];
+    };
+}
+/** The agent-scoped context's waterfall surface the router uses. */
+interface AgentScopedContext {
+    on(event: "agent/request", listener: (payload: Record<string, unknown>, next: () => Promise<Record<string, unknown>>) => Promise<Record<string, unknown>>, options?: {
+        prepend?: boolean;
+    }): () => void;
+}
+/** Host-plane surface the router consumes (events, prompt registry, skills). */
+interface HarnessContext {
+    on(event: "agent/created", listener: (payload: {
+        agent: AgentLike;
+    }) => void): () => void;
+    on(event: "agent/disposed", listener: (agent: unknown) => void): () => void;
+    systemPrompt: {
+        section(section: {
+            name: string;
+            order: number;
+            text: string;
+        }): () => void;
+    };
+    skills: {
+        register(skill: {
+            name: string;
+            description: string;
+            whenToUse?: string;
+            content: string;
+            source: string;
+        }): () => void;
+    };
+}
 /**
  * Cordis service: per-agent request routing plus the convention surface.
  */
 declare class ModelRouter extends Service {
     static inject: string[];
     config: RouterConfig;
+    /** Currently authoritative config; swapped by the settings section when one is mounted. */
+    source: () => RouterConfig;
+    /** Host-plane surface the router consumes (events, prompt registry, skills). */
+    harness: HarnessContext;
+    /** Disposer of the currently registered prompt section, if any. */
+    promptDispose?: () => void;
+    /** Disposer of the currently registered skill, if any. */
+    skillDispose?: () => void;
     constructor(ctx: Context, rawConfig?: unknown);
+    /**
+     * Register the convention surface (prompt section + skill) for `cfg`,
+     * replacing whatever is currently registered. Called on construction and
+     * after every committed settings change; while `enabled` is false, nothing
+     * stays registered.
+     */
+    private render;
 }
 
-export { type AgentRole, Config, type ModelRoute, ModelRouter, ROW_ID, type ReasoningEffort, type RouterConfig, type RoutingMode, SKILL_CONTENT, SKILL_DESCRIPTION, SKILL_NAME, SKILL_WHEN_TO_USE, ModelRouter as default, name, roleFor, routeFor };
+export { type AgentRole, Config, type ModelRoute, ModelRouter, ROW_ID, type ReasoningEffort, type RouterConfig, type RoutingMode, SETTINGS_NS, SKILL_CONTENT, SKILL_DESCRIPTION, SKILL_NAME, SKILL_WHEN_TO_USE, ModelRouter as default, effortFor, name, recentStepsHadError, roleFor, routeFor };

package/lib/index.js

@@ -2,8 +2,10 @@
 import { Service } from "@deepseek-ai/cordis";
 import z from "@deepseek-ai/schemastery";
 import { foldPlanMode } from "@deepseek-ai/dsh-plan-mode";
+import { installSettingsSection, settingsNamespace } from "@deepseek-ai/dsh-settings";
 
 // src/policy.ts
+var DEFAULT_RECOVERY_STEPS = 2;
 function roleFor(agent) {
   const options = agent?.options;
   const depth = options?.subagentDepth;
@@ -19,6 +21,29 @@ function routeFor(agent, config, planModeActive = false) {
   if (config.mode === "plan" && !planModeActive) return config.executor;
   return config.planner;
 }
+function recentStepsHadError(events, recoverySteps = DEFAULT_RECOVERY_STEPS) {
+  if (!Array.isArray(events) || recoverySteps <= 0) return false;
+  const seen = /* @__PURE__ */ new Set();
+  let steps = 0;
+  for (let i = events.length - 1; i >= 0; i -= 1) {
+    const event = events[i];
+    if (event?.type !== "tool/result" || event.data === void 0) continue;
+    const key = `${event.data.turn}:${event.data.step}`;
+    if (!seen.has(key)) {
+      if (steps >= recoverySteps) break;
+      seen.add(key);
+      steps += 1;
+    }
+    if (event.data.error !== void 0 && event.data.error !== null) return true;
+  }
+  return false;
+}
+function effortFor(route, events) {
+  if (route.escalateOnError === true && recentStepsHadError(events, route.recoverySteps)) {
+    return route.escalateTo ?? route.reasoningEffort;
+  }
+  return route.reasoningEffort;
+}
 
 // src/index.ts
 var name = "model-router";
@@ -26,7 +51,10 @@ var ModelRouteSchema = z.object({
   provider: z.string().min(1),
   model: z.string().min(1),
   reasoningEffort: z.union(["off", "low", "high", "max"]),
-  maxTokens: z.number().min(1)
+  maxTokens: z.number().min(1),
+  escalateOnError: z.boolean(),
+  escalateTo: z.union(["off", "low", "high", "max"]),
+  recoverySteps: z.number().min(1)
 });
 var Config = z.object({
   planner: ModelRouteSchema.default({
@@ -38,15 +66,18 @@ var Config = z.object({
     model: "deepseek-v4-flash"
   }),
   mode: z.union(["strict", "plan"]).default("strict"),
+  enabled: z.boolean().default(true),
   promptSection: z.boolean().default(true),
   skill: z.boolean().default(true)
 });
+var SETTINGS_NS = settingsNamespace("model-router");
 function resolveConfig(raw) {
   const parsed = Config(raw ?? {});
   return {
     planner: parsed.planner,
     executor: parsed.executor,
     mode: parsed.mode,
+    enabled: parsed.enabled,
     promptSection: parsed.promptSection,
     skill: parsed.skill
   };
@@ -98,44 +129,86 @@ function isPlanModeActive(agent) {
 var ModelRouter = class extends Service {
   static inject = ["skills", "systemPrompt"];
   config;
+  /** Currently authoritative config; swapped by the settings section when one is mounted. */
+  source;
+  /** Host-plane surface the router consumes (events, prompt registry, skills). */
+  harness;
+  /** Disposer of the currently registered prompt section, if any. */
+  promptDispose;
+  /** Disposer of the currently registered skill, if any. */
+  skillDispose;
   constructor(ctx, rawConfig = {}) {
     super(ctx, "modelRouter");
     this.config = resolveConfig(rawConfig);
-    const harness = ctx;
-    harness.on("agent/created", ({ agent }) => {
+    this.source = () => this.config;
+    this.harness = ctx;
+    this.harness.on("agent/created", ({ agent }) => {
       const dispose = agent.ctx.on(
         "agent/request",
         async (payload, next) => {
           const resolved = await next();
-          const route = routeFor(agent, this.config, isPlanModeActive(agent));
+          const cfg = this.source();
+          if (!cfg.enabled) return resolved;
+          const route = routeFor(agent, cfg, isPlanModeActive(agent));
           if (route === void 0) return resolved;
           const stamped = {
             ...resolved,
             provider: route.provider,
             model: route.model
           };
-          if (route.reasoningEffort !== void 0) stamped.reasoningEffort = route.reasoningEffort;
           if (route.maxTokens !== void 0) stamped.maxTokens = route.maxTokens;
+          const effort = effortFor(route, agent.session?.events);
+          if (effort !== void 0) stamped.reasoningEffort = effort;
           return stamped;
         },
         { prepend: true }
       );
-      harness.on("agent/disposed", (disposed) => {
+      this.harness.on("agent/disposed", (disposed) => {
         if (disposed === agent) dispose();
       });
     });
-    if (this.config.promptSection) {
-      harness.systemPrompt.section({
+    this.render(this.config);
+    installSettingsSection(
+      ctx,
+      SETTINGS_NS,
+      Config,
+      this.config,
+      {
+        setSource: (current) => {
+          this.source = current;
+        },
+        onChange: () => this.render(this.source())
+      }
+    );
+  }
+  /**
+   * Register the convention surface (prompt section + skill) for `cfg`,
+   * replacing whatever is currently registered. Called on construction and
+   * after every committed settings change; while `enabled` is false, nothing
+   * stays registered.
+   */
+  render(cfg) {
+    if (this.promptDispose) {
+      this.promptDispose();
+      this.promptDispose = void 0;
+    }
+    if (this.skillDispose) {
+      this.skillDispose();
+      this.skillDispose = void 0;
+    }
+    if (!cfg.enabled) return;
+    if (cfg.promptSection) {
+      this.promptDispose = this.harness.systemPrompt.section({
         name: ROW_ID,
         order: SECTION_ORDER,
-        text: SECTION_TEXT.replaceAll("{PLANNER_MODEL}", this.config.planner.model).replaceAll(
+        text: SECTION_TEXT.replaceAll("{PLANNER_MODEL}", cfg.planner.model).replaceAll(
           "{EXECUTOR_MODEL}",
-          this.config.executor.model
+          cfg.executor.model
         )
       });
     }
-    if (this.config.skill) {
-      harness.skills.register({
+    if (cfg.skill) {
+      this.skillDispose = this.harness.skills.register({
         name: SKILL_NAME,
         description: SKILL_DESCRIPTION,
         whenToUse: SKILL_WHEN_TO_USE,
@@ -149,12 +222,15 @@ export {
   Config,
   ModelRouter,
   ROW_ID,
+  SETTINGS_NS,
   SKILL_CONTENT,
   SKILL_DESCRIPTION,
   SKILL_NAME,
   SKILL_WHEN_TO_USE,
   ModelRouter as default,
+  effortFor,
   name,
+  recentStepsHadError,
   roleFor,
   routeFor
 };

package/lib/index.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/index.ts","../src/policy.ts"],"sourcesContent":["/**\n * dsh-model-router: role-based model routing for the DeepSeek Harness.\n *\n * The planner (the session's root agent) runs on `deepseek-v4-pro`; delegated\n * executor subagents run on `deepseek-v4-flash`. Enforcement is a per-agent\n * `agent/request` rewrite registered when the agent is created, so it applies\n * in every mode (web / headless / tui) and every agent preset, including\n * subagents the delegation tools create.\n *\n * Each role route may also pin `reasoningEffort` and `maxTokens`; when set,\n * they override the session's selection for that role. A `mode` switch lets a\n * deployment reserve the planner route for actual planning.\n *\n * The plugin also publishes:\n *  - a system-prompt section stating the planner/executor convention, and\n *  - the `pro-flash-routing` skill teaching the agent to plan itself and\n *    delegate code execution to flash subagents.\n *\n * @module dsh-model-router\n */\nimport { Context, Service } from \"@deepseek-ai/cordis\";\nimport z from \"@deepseek-ai/schemastery\";\nimport { foldPlanMode } from \"@deepseek-ai/dsh-plan-mode\";\nimport { roleFor, routeFor, type RouterConfig } from \"./policy.js\";\n\n/** Plugin row id; the bundle patch inserts it under this id. */\nconst name = \"model-router\";\n\n/** One provider/model pair, with defaults and optional effort/token caps. */\nconst ModelRouteSchema = z.object({\n  provider: z.string().min(1),\n  model: z.string().min(1),\n  reasoningEffort: z.union([\"off\", \"low\", \"high\", \"max\"]),\n  maxTokens: z.number().min(1),\n});\n\n/** The plugin's public config, validated at row load. */\nconst Config = z.object({\n  planner: ModelRouteSchema.default({\n    provider: \"deepseek-official\",\n    model: \"deepseek-v4-pro\",\n  } as never),\n  executor: ModelRouteSchema.default({\n    provider: \"deepseek-official\",\n    model: \"deepseek-v4-flash\",\n  } as never),\n  mode: z.union([\"strict\", \"plan\"]).default(\"strict\"),\n  promptSection: z.boolean().default(true),\n  skill: z.boolean().default(true),\n});\n\n/**\n * Resolve raw row config into the internal shape, failing loud on garbage.\n * Schemastery schemas are callable: invoking validates and applies defaults.\n * @param raw - the row's config object.\n * @returns the validated RouterConfig.\n */\nfunction resolveConfig(raw: unknown): RouterConfig {\n  const parsed = Config(raw ?? {});\n  return {\n    planner: parsed.planner,\n    executor: parsed.executor,\n    mode: parsed.mode,\n    promptSection: parsed.promptSection,\n    skill: parsed.skill,\n  };\n}\n\n/**\n * Always-on guidance section. Negative order renders before the persona, so\n * the convention is established before the agent's identity line.\n */\nconst SECTION_ORDER = -50;\n\nconst SECTION_TEXT = `Model routing is role-based. Planning runs on {PLANNER_MODEL}; implementation runs on {EXECUTOR_MODEL}. You are the root agent: plan, design, review subagent output, and write the final answer here. Delegate implementation — writing code, running commands, builds, tests — to subagents with complete, self-contained prompts, preferring background delegation for independent work. Keep plans and replies concise. Do not hand-write large amounts of code or run long executions here; delegate instead.`;\n\nconst SKILL_NAME = \"pro-flash-routing\";\n\nconst SKILL_DESCRIPTION =\n  \"Route planning and code execution across models: plan on the pro planner agent, delegate implementation to flash executor subagents.\";\n\nconst SKILL_WHEN_TO_USE = `Use when a task combines planning and implementation: before writing code, after a plan is approved, when delegating execution work, or when the user asks about the pro/flash routing convention.`;\n\nconst SKILL_CONTENT = `# Pro planner / Flash executor routing\n\nThis session routes models by role:\n\n- **Planner (this agent)** — \\`deepseek-v4-pro\\`. Planning, design decisions, reviewing delegated output, and user-facing synthesis happen here.\n- **Executors (every subagent)** — \\`deepseek-v4-flash\\`. Implementation work happens there: writing code, running commands, builds, and tests. The harness forces the model automatically; you do not select it.\n\n## Working rhythm\n\n1. **Plan here.** Explore, decide the approach, and (when plan mode is on) submit the plan with \\`exit_plan_mode\\`. The plan stays on this agent.\n2. **Delegate the execution.** Once a plan is approved, hand each self-contained chunk of implementation to a subagent with a complete prompt: exact files to touch, the change to make, and how to verify. Subagents are automatically routed to \\`deepseek-v4-flash\\`, so keep them execution-focused: give them the decision, not the decision to make.\n3. **Review here.** Read the subagent's result on this agent, verify it yourself (tests, diffs, logs), and iterate with follow-up messages to the same subagent when available.\n4. **Report here.** Summaries, plans, and answers to the user come from this agent.\n\n## Keep this agent's context lean\n\nInput tokens are the expensive part of the planner. Don't re-read large files or full transcripts on this agent — trust the subagent's final report. Prefer targeted reads (offset/limit) over whole files. When the context grows, compact rather than re-sending everything.\n\n## Delegation guidelines\n\n- Start independent delegations together in one assistant message and continue useful work while they run (background mode by default).\n- Prefer \\`subagent\\` for self-contained work and \\`workflow\\` when many independent pieces need fan-out; their workers run on flash as well.\n- Do not delegate design: subagents execute decisions already made.\n- If a subagent's task grows into design work, pull it back to this agent and re-delegate the narrowed execution.\n\n## Verification\n\n- Executor output was produced by \\`deepseek-v4-flash\\`; planner output by \\`deepseek-v4-pro\\`. If you need to confirm, check the session log's model metadata.\n- If routing ever looks wrong, the \\`model-router\\` plugin row in the profile composition is the single place that owns it.`;\n\n/** The plugin row id the bundle patch must insert. */\nconst ROW_ID = \"model-router\";\n\n/** Minimal structural view of the live agent object the router reads. */\ninterface AgentLike {\n  ctx: AgentScopedContext;\n  options?: { subagentDepth?: number };\n  session?: { header?: { origin?: string }; events?: unknown[] };\n}\n\n/** The agent-scoped context's waterfall surface the router uses. */\ninterface AgentScopedContext {\n  on(\n    event: \"agent/request\",\n    listener: (\n      payload: Record<string, unknown>,\n      next: () => Promise<Record<string, unknown>>,\n    ) => Promise<Record<string, unknown>>,\n    options?: { prepend?: boolean },\n  ): () => void;\n}\n\n/** Host-plane surface the router consumes (events, prompt registry, skills). */\ninterface HarnessContext {\n  on(\n    event: \"agent/created\",\n    listener: (payload: { agent: AgentLike }) => void,\n  ): () => void;\n  on(event: \"agent/disposed\", listener: (agent: unknown) => void): () => void;\n  systemPrompt: {\n    section(section: { name: string; order: number; text: string }): unknown;\n  };\n  skills: {\n    register(skill: {\n      name: string;\n      description: string;\n      whenToUse?: string;\n      content: string;\n      source: string;\n    }): unknown;\n  };\n}\n\n/** Fold plan-mode state for an agent without trusting the agent's exact shape. */\nfunction isPlanModeActive(agent: AgentLike): boolean {\n  const events = agent.session?.events;\n  if (!Array.isArray(events)) return false;\n  try {\n    return foldPlanMode(events as Parameters<typeof foldPlanMode>[0]);\n  } catch {\n    return false;\n  }\n}\n\n/**\n * Cordis service: per-agent request routing plus the convention surface.\n */\nclass ModelRouter extends Service {\n  static inject = [\"skills\", \"systemPrompt\"];\n\n  config: RouterConfig;\n\n  constructor(ctx: Context, rawConfig: unknown = {}) {\n    super(ctx, \"modelRouter\");\n    this.config = resolveConfig(rawConfig);\n    const harness = ctx as unknown as HarnessContext;\n\n    // Every agent that gets created — root sessions, delegation children,\n    // workflow workers, ralph rounds — passes through here.\n    harness.on(\"agent/created\", ({ agent }) => {\n      // `prepend` puts this listener OUTERMOST in the `agent/request`\n      // waterfall: the harness's model-selection listener runs inside it, so\n      // this rewrite is applied LAST and wins over the session's selected\n      // model (which dsh-base defaults to deepseek-v4-flash and the user\n      // settings or UI can change).\n      const dispose = agent.ctx.on(\n        \"agent/request\",\n        async (payload, next) => {\n          const resolved = await next();\n          const route = routeFor(agent, this.config, isPlanModeActive(agent));\n          if (route === undefined) return resolved;\n          const stamped: Record<string, unknown> = {\n            ...resolved,\n            provider: route.provider,\n            model: route.model,\n          };\n          if (route.reasoningEffort !== undefined) stamped.reasoningEffort = route.reasoningEffort;\n          if (route.maxTokens !== undefined) stamped.maxTokens = route.maxTokens;\n          return stamped;\n        },\n        { prepend: true },\n      );\n      harness.on(\"agent/disposed\", (disposed) => {\n        if (disposed === agent) dispose();\n      });\n    });\n\n    if (this.config.promptSection) {\n      harness.systemPrompt.section({\n        name: ROW_ID,\n        order: SECTION_ORDER,\n        text: SECTION_TEXT.replaceAll(\"{PLANNER_MODEL}\", this.config.planner.model).replaceAll(\n          \"{EXECUTOR_MODEL}\",\n          this.config.executor.model,\n        ),\n      });\n    }\n\n    if (this.config.skill) {\n      harness.skills.register({\n        name: SKILL_NAME,\n        description: SKILL_DESCRIPTION,\n        whenToUse: SKILL_WHEN_TO_USE,\n        content: SKILL_CONTENT,\n        source: \"runtime\",\n      });\n    }\n  }\n}\n\nexport {\n  Config,\n  ModelRouter,\n  ModelRouter as default,\n  name,\n  ROW_ID,\n  SKILL_CONTENT,\n  SKILL_DESCRIPTION,\n  SKILL_NAME,\n  SKILL_WHEN_TO_USE,\n  roleFor,\n  routeFor,\n};\nexport type {\n  AgentRole,\n  ModelRoute,\n  ReasoningEffort,\n  RoutingMode,\n  RouterConfig,\n} from \"./policy.js\";\n","/**\n * Pure routing policy for dsh-model-router: which model each agent role gets.\n * Kept free of Cordis imports so the policy is trivially unit-testable.\n * @module dsh-model-router/policy\n */\n\n/** Reasoning-effort levels a route may pin (mirrors the harness vocabulary). */\nexport type ReasoningEffort = \"off\" | \"low\" | \"high\" | \"max\";\n\n/**\n * How the router treats the root agent.\n * - `strict`: the root agent is always the planner (pro).\n * - `plan`: the root agent is pro only while plan mode is active; otherwise it\n *   falls back to the executor route, reserving pro for real planning.\n */\nexport type RoutingMode = \"strict\" | \"plan\";\n\n/** One route: a provider/model pair stamped onto an agent request. */\nexport interface ModelRoute {\n  provider: string;\n  model: string;\n  /**\n   * Optional reasoning-effort override. When omitted, the request inherits the\n   * session's own selection; when set, the router pins it for that role.\n   */\n  reasoningEffort?: ReasoningEffort;\n  /** Optional output-token cap for the role; omitted means inherit. */\n  maxTokens?: number;\n}\n\n/** The two roles the router distinguishes. */\nexport type AgentRole = \"planner\" | \"executor\";\n\n/** Resolved router configuration: one route per role plus routing mode. */\nexport interface RouterConfig {\n  planner: ModelRoute;\n  executor: ModelRoute;\n  mode: RoutingMode;\n  promptSection: boolean;\n  skill: boolean;\n}\n\n/**\n * Classify an agent as planner or executor.\n *\n * The main (root) agent of a session is the planner. Every agent created as a\n * delegation child — `subagent`, `subagent_fork`, workflow workers, ralph\n * rounds — is an executor. The harness stamps two durable facts on children:\n * `options.subagentDepth` (>= 1) and the session header `origin: \"subagent\"`.\n *\n * @param agent - the live agent (any subset of the runtime shape).\n * @returns the role the agent should be routed as.\n */\nexport function roleFor(agent: unknown): AgentRole {\n  const options = (agent as { options?: unknown })?.options;\n  const depth = (options as { subagentDepth?: unknown })?.subagentDepth;\n  if (typeof depth === \"number\" && depth >= 1) return \"executor\";\n  const session = (agent as { session?: unknown })?.session;\n  const origin = (session as { header?: unknown })?.header\n    ? ((session as { header: { origin?: unknown } }).header.origin)\n    : undefined;\n  if (origin === \"subagent\") return \"executor\";\n  return \"planner\";\n}\n\n/**\n * Resolve the route for one agent.\n * @param agent - the live agent.\n * @param config - the resolved router configuration.\n * @param planModeActive - whether plan mode is currently folded active for the\n *   agent's session; consulted only in `plan` routing mode.\n * @returns the model route to stamp, or `undefined` to leave the request alone.\n */\nexport function routeFor(\n  agent: unknown,\n  config: RouterConfig,\n  planModeActive = false,\n): ModelRoute | undefined {\n  const role = roleFor(agent);\n  if (role === \"executor\") return config.executor;\n  // Root agent. In `plan` mode, reserve the planner route for actual planning;\n  // otherwise the root falls back to the executor route.\n  if (config.mode === \"plan\" && !planModeActive) return config.executor;\n  return config.planner;\n}\n"],"mappings":";AAoBA,SAAkB,eAAe;AACjC,OAAO,OAAO;AACd,SAAS,oBAAoB;;;AC+BtB,SAAS,QAAQ,OAA2B;AACjD,QAAM,UAAW,OAAiC;AAClD,QAAM,QAAS,SAAyC;AACxD,MAAI,OAAO,UAAU,YAAY,SAAS,EAAG,QAAO;AACpD,QAAM,UAAW,OAAiC;AAClD,QAAM,SAAU,SAAkC,SAC5C,QAA6C,OAAO,SACtD;AACJ,MAAI,WAAW,WAAY,QAAO;AAClC,SAAO;AACT;AAUO,SAAS,SACd,OACA,QACA,iBAAiB,OACO;AACxB,QAAM,OAAO,QAAQ,KAAK;AAC1B,MAAI,SAAS,WAAY,QAAO,OAAO;AAGvC,MAAI,OAAO,SAAS,UAAU,CAAC,eAAgB,QAAO,OAAO;AAC7D,SAAO,OAAO;AAChB;;;AD1DA,IAAM,OAAO;AAGb,IAAM,mBAAmB,EAAE,OAAO;AAAA,EAChC,UAAU,EAAE,OAAO,EAAE,IAAI,CAAC;AAAA,EAC1B,OAAO,EAAE,OAAO,EAAE,IAAI,CAAC;AAAA,EACvB,iBAAiB,EAAE,MAAM,CAAC,OAAO,OAAO,QAAQ,KAAK,CAAC;AAAA,EACtD,WAAW,EAAE,OAAO,EAAE,IAAI,CAAC;AAC7B,CAAC;AAGD,IAAM,SAAS,EAAE,OAAO;AAAA,EACtB,SAAS,iBAAiB,QAAQ;AAAA,IAChC,UAAU;AAAA,IACV,OAAO;AAAA,EACT,CAAU;AAAA,EACV,UAAU,iBAAiB,QAAQ;AAAA,IACjC,UAAU;AAAA,IACV,OAAO;AAAA,EACT,CAAU;AAAA,EACV,MAAM,EAAE,MAAM,CAAC,UAAU,MAAM,CAAC,EAAE,QAAQ,QAAQ;AAAA,EAClD,eAAe,EAAE,QAAQ,EAAE,QAAQ,IAAI;AAAA,EACvC,OAAO,EAAE,QAAQ,EAAE,QAAQ,IAAI;AACjC,CAAC;AAQD,SAAS,cAAc,KAA4B;AACjD,QAAM,SAAS,OAAO,OAAO,CAAC,CAAC;AAC/B,SAAO;AAAA,IACL,SAAS,OAAO;AAAA,IAChB,UAAU,OAAO;AAAA,IACjB,MAAM,OAAO;AAAA,IACb,eAAe,OAAO;AAAA,IACtB,OAAO,OAAO;AAAA,EAChB;AACF;AAMA,IAAM,gBAAgB;AAEtB,IAAM,eAAe;AAErB,IAAM,aAAa;AAEnB,IAAM,oBACJ;AAEF,IAAM,oBAAoB;AAE1B,IAAM,gBAAgB;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AA+BtB,IAAM,SAAS;AA2Cf,SAAS,iBAAiB,OAA2B;AACnD,QAAM,SAAS,MAAM,SAAS;AAC9B,MAAI,CAAC,MAAM,QAAQ,MAAM,EAAG,QAAO;AACnC,MAAI;AACF,WAAO,aAAa,MAA4C;AAAA,EAClE,QAAQ;AACN,WAAO;AAAA,EACT;AACF;AAKA,IAAM,cAAN,cAA0B,QAAQ;AAAA,EAChC,OAAO,SAAS,CAAC,UAAU,cAAc;AAAA,EAEzC;AAAA,EAEA,YAAY,KAAc,YAAqB,CAAC,GAAG;AACjD,UAAM,KAAK,aAAa;AACxB,SAAK,SAAS,cAAc,SAAS;AACrC,UAAM,UAAU;AAIhB,YAAQ,GAAG,iBAAiB,CAAC,EAAE,MAAM,MAAM;AAMzC,YAAM,UAAU,MAAM,IAAI;AAAA,QACxB;AAAA,QACA,OAAO,SAAS,SAAS;AACvB,gBAAM,WAAW,MAAM,KAAK;AAC5B,gBAAM,QAAQ,SAAS,OAAO,KAAK,QAAQ,iBAAiB,KAAK,CAAC;AAClE,cAAI,UAAU,OAAW,QAAO;AAChC,gBAAM,UAAmC;AAAA,YACvC,GAAG;AAAA,YACH,UAAU,MAAM;AAAA,YAChB,OAAO,MAAM;AAAA,UACf;AACA,cAAI,MAAM,oBAAoB,OAAW,SAAQ,kBAAkB,MAAM;AACzE,cAAI,MAAM,cAAc,OAAW,SAAQ,YAAY,MAAM;AAC7D,iBAAO;AAAA,QACT;AAAA,QACA,EAAE,SAAS,KAAK;AAAA,MAClB;AACA,cAAQ,GAAG,kBAAkB,CAAC,aAAa;AACzC,YAAI,aAAa,MAAO,SAAQ;AAAA,MAClC,CAAC;AAAA,IACH,CAAC;AAED,QAAI,KAAK,OAAO,eAAe;AAC7B,cAAQ,aAAa,QAAQ;AAAA,QAC3B,MAAM;AAAA,QACN,OAAO;AAAA,QACP,MAAM,aAAa,WAAW,mBAAmB,KAAK,OAAO,QAAQ,KAAK,EAAE;AAAA,UAC1E;AAAA,UACA,KAAK,OAAO,SAAS;AAAA,QACvB;AAAA,MACF,CAAC;AAAA,IACH;AAEA,QAAI,KAAK,OAAO,OAAO;AACrB,cAAQ,OAAO,SAAS;AAAA,QACtB,MAAM;AAAA,QACN,aAAa;AAAA,QACb,WAAW;AAAA,QACX,SAAS;AAAA,QACT,QAAQ;AAAA,MACV,CAAC;AAAA,IACH;AAAA,EACF;AACF;","names":[]}
+{"version":3,"sources":["../src/index.ts","../src/policy.ts"],"sourcesContent":["/**\n * dsh-model-router: role-based model routing for the DeepSeek Harness.\n *\n * The planner (the session's root agent) runs on `deepseek-v4-pro`; delegated\n * executor subagents run on `deepseek-v4-flash`. Enforcement is a per-agent\n * `agent/request` rewrite registered when the agent is created, so it applies\n * in every mode (web / headless / tui) and every agent preset, including\n * subagents the delegation tools create.\n *\n * Each role route may also pin `reasoningEffort` and `maxTokens`; when set,\n * they override the session's selection for that role. A `mode` switch lets a\n * deployment reserve the planner route for actual planning.\n *\n * The plugin also publishes:\n *  - a system-prompt section stating the planner/executor convention, and\n *  - the `pro-flash-routing` skill teaching the agent to plan itself and\n *    delegate code execution to flash subagents.\n *\n * @module dsh-model-router\n */\nimport { Context, Service } from \"@deepseek-ai/cordis\";\nimport z from \"@deepseek-ai/schemastery\";\nimport { foldPlanMode } from \"@deepseek-ai/dsh-plan-mode\";\nimport { installSettingsSection, settingsNamespace } from \"@deepseek-ai/dsh-settings\";\nimport { effortFor, recentStepsHadError, roleFor, routeFor, type RouterConfig } from \"./policy.js\";\n\n/** Plugin row id; the bundle patch inserts it under this id. */\nconst name = \"model-router\";\n\n/** One provider/model pair, with defaults and optional effort/token caps. */\nconst ModelRouteSchema = z.object({\n  provider: z.string().min(1),\n  model: z.string().min(1),\n  reasoningEffort: z.union([\"off\", \"low\", \"high\", \"max\"]),\n  maxTokens: z.number().min(1),\n  escalateOnError: z.boolean(),\n  escalateTo: z.union([\"off\", \"low\", \"high\", \"max\"]),\n  recoverySteps: z.number().min(1),\n});\n\n/** The plugin's public config, validated at row load. */\nconst Config = z.object({\n  planner: ModelRouteSchema.default({\n    provider: \"deepseek-official\",\n    model: \"deepseek-v4-pro\",\n  } as never),\n  executor: ModelRouteSchema.default({\n    provider: \"deepseek-official\",\n    model: \"deepseek-v4-flash\",\n  } as never),\n  mode: z.union([\"strict\", \"plan\"]).default(\"strict\"),\n  enabled: z.boolean().default(true),\n  promptSection: z.boolean().default(true),\n  skill: z.boolean().default(true),\n});\n\n/** Settings namespace the live on/off toggle lives under (settings.yaml). */\nconst SETTINGS_NS = settingsNamespace(\"model-router\");\n\n/**\n * Resolve raw row config into the internal shape, failing loud on garbage.\n * Schemastery schemas are callable: invoking validates and applies defaults.\n * @param raw - the row's config object.\n * @returns the validated RouterConfig.\n */\nfunction resolveConfig(raw: unknown): RouterConfig {\n  const parsed = Config(raw ?? {});\n  return {\n    planner: parsed.planner,\n    executor: parsed.executor,\n    mode: parsed.mode,\n    enabled: parsed.enabled,\n    promptSection: parsed.promptSection,\n    skill: parsed.skill,\n  };\n}\n\n/**\n * Always-on guidance section. Negative order renders before the persona, so\n * the convention is established before the agent's identity line.\n */\nconst SECTION_ORDER = -50;\n\nconst SECTION_TEXT = `Model routing is role-based. Planning runs on {PLANNER_MODEL}; implementation runs on {EXECUTOR_MODEL}. You are the root agent: plan, design, review subagent output, and write the final answer here. Delegate implementation — writing code, running commands, builds, tests — to subagents with complete, self-contained prompts, preferring background delegation for independent work. Keep plans and replies concise. Do not hand-write large amounts of code or run long executions here; delegate instead.`;\n\nconst SKILL_NAME = \"pro-flash-routing\";\n\nconst SKILL_DESCRIPTION =\n  \"Route planning and code execution across models: plan on the pro planner agent, delegate implementation to flash executor subagents.\";\n\nconst SKILL_WHEN_TO_USE = `Use when a task combines planning and implementation: before writing code, after a plan is approved, when delegating execution work, or when the user asks about the pro/flash routing convention.`;\n\nconst SKILL_CONTENT = `# Pro planner / Flash executor routing\n\nThis session routes models by role:\n\n- **Planner (this agent)** — \\`deepseek-v4-pro\\`. Planning, design decisions, reviewing delegated output, and user-facing synthesis happen here.\n- **Executors (every subagent)** — \\`deepseek-v4-flash\\`. Implementation work happens there: writing code, running commands, builds, and tests. The harness forces the model automatically; you do not select it.\n\n## Working rhythm\n\n1. **Plan here.** Explore, decide the approach, and (when plan mode is on) submit the plan with \\`exit_plan_mode\\`. The plan stays on this agent.\n2. **Delegate the execution.** Once a plan is approved, hand each self-contained chunk of implementation to a subagent with a complete prompt: exact files to touch, the change to make, and how to verify. Subagents are automatically routed to \\`deepseek-v4-flash\\`, so keep them execution-focused: give them the decision, not the decision to make.\n3. **Review here.** Read the subagent's result on this agent, verify it yourself (tests, diffs, logs), and iterate with follow-up messages to the same subagent when available.\n4. **Report here.** Summaries, plans, and answers to the user come from this agent.\n\n## Keep this agent's context lean\n\nInput tokens are the expensive part of the planner. Don't re-read large files or full transcripts on this agent — trust the subagent's final report. Prefer targeted reads (offset/limit) over whole files. When the context grows, compact rather than re-sending everything.\n\n## Delegation guidelines\n\n- Start independent delegations together in one assistant message and continue useful work while they run (background mode by default).\n- Prefer \\`subagent\\` for self-contained work and \\`workflow\\` when many independent pieces need fan-out; their workers run on flash as well.\n- Do not delegate design: subagents execute decisions already made.\n- If a subagent's task grows into design work, pull it back to this agent and re-delegate the narrowed execution.\n\n## Verification\n\n- Executor output was produced by \\`deepseek-v4-flash\\`; planner output by \\`deepseek-v4-pro\\`. If you need to confirm, check the session log's model metadata.\n- If routing ever looks wrong, the \\`model-router\\` plugin row in the profile composition is the single place that owns it.`;\n\n/** The plugin row id the bundle patch must insert. */\nconst ROW_ID = \"model-router\";\n\n/** Minimal structural view of the live agent object the router reads. */\ninterface AgentLike {\n  ctx: AgentScopedContext;\n  options?: { subagentDepth?: number };\n  session?: { header?: { origin?: string }; events?: unknown[] };\n}\n\n/** The agent-scoped context's waterfall surface the router uses. */\ninterface AgentScopedContext {\n  on(\n    event: \"agent/request\",\n    listener: (\n      payload: Record<string, unknown>,\n      next: () => Promise<Record<string, unknown>>,\n    ) => Promise<Record<string, unknown>>,\n    options?: { prepend?: boolean },\n  ): () => void;\n}\n\n/** Host-plane surface the router consumes (events, prompt registry, skills). */\ninterface HarnessContext {\n  on(\n    event: \"agent/created\",\n    listener: (payload: { agent: AgentLike }) => void,\n  ): () => void;\n  on(event: \"agent/disposed\", listener: (agent: unknown) => void): () => void;\n  systemPrompt: {\n    section(section: { name: string; order: number; text: string }): () => void;\n  };\n  skills: {\n    register(skill: {\n      name: string;\n      description: string;\n      whenToUse?: string;\n      content: string;\n      source: string;\n    }): () => void;\n  };\n}\n\n/** Fold plan-mode state for an agent without trusting the agent's exact shape. */\nfunction isPlanModeActive(agent: AgentLike): boolean {\n  const events = agent.session?.events;\n  if (!Array.isArray(events)) return false;\n  try {\n    return foldPlanMode(events as Parameters<typeof foldPlanMode>[0]);\n  } catch {\n    return false;\n  }\n}\n\n/**\n * Cordis service: per-agent request routing plus the convention surface.\n */\nclass ModelRouter extends Service {\n  static inject = [\"skills\", \"systemPrompt\"];\n\n  config: RouterConfig;\n\n  /** Currently authoritative config; swapped by the settings section when one is mounted. */\n  source: () => RouterConfig;\n\n  /** Host-plane surface the router consumes (events, prompt registry, skills). */\n  harness: HarnessContext;\n\n  /** Disposer of the currently registered prompt section, if any. */\n  promptDispose?: () => void;\n\n  /** Disposer of the currently registered skill, if any. */\n  skillDispose?: () => void;\n\n  constructor(ctx: Context, rawConfig: unknown = {}) {\n    super(ctx, \"modelRouter\");\n    this.config = resolveConfig(rawConfig);\n    this.source = () => this.config;\n    this.harness = ctx as unknown as HarnessContext;\n\n    // Every agent that gets created — root sessions, delegation children,\n    // workflow workers, ralph rounds — passes through here.\n    this.harness.on(\"agent/created\", ({ agent }) => {\n      // `prepend` puts this listener OUTERMOST in the `agent/request`\n      // waterfall: the harness's model-selection listener runs inside it, so\n      // this rewrite is applied LAST and wins over the session's selected\n      // model (which dsh-base defaults to deepseek-v4-flash and the user\n      // settings or UI can change).\n      const dispose = agent.ctx.on(\n        \"agent/request\",\n        async (payload, next) => {\n          const resolved = await next();\n          const cfg = this.source();\n          if (!cfg.enabled) return resolved;\n          const route = routeFor(agent, cfg, isPlanModeActive(agent));\n          if (route === undefined) return resolved;\n          const stamped: Record<string, unknown> = {\n            ...resolved,\n            provider: route.provider,\n            model: route.model,\n          };\n          if (route.maxTokens !== undefined) stamped.maxTokens = route.maxTokens;\n          const effort = effortFor(route, agent.session?.events);\n          if (effort !== undefined) stamped.reasoningEffort = effort;\n          return stamped;\n        },\n        { prepend: true },\n      );\n      this.harness.on(\"agent/disposed\", (disposed) => {\n        if (disposed === agent) dispose();\n      });\n    });\n\n    // Register the convention surface from the composition config, then let\n    // the settings section (when mounted) take over as the live source.\n    this.render(this.config);\n    installSettingsSection(\n      ctx,\n      SETTINGS_NS,\n      Config,\n      this.config as unknown as ReturnType<typeof Config>,\n      {\n        setSource: (current) => {\n          this.source = current;\n        },\n        onChange: () => this.render(this.source()),\n      },\n    );\n  }\n\n  /**\n   * Register the convention surface (prompt section + skill) for `cfg`,\n   * replacing whatever is currently registered. Called on construction and\n   * after every committed settings change; while `enabled` is false, nothing\n   * stays registered.\n   */\n  private render(cfg: RouterConfig): void {\n    if (this.promptDispose) {\n      this.promptDispose();\n      this.promptDispose = undefined;\n    }\n    if (this.skillDispose) {\n      this.skillDispose();\n      this.skillDispose = undefined;\n    }\n    if (!cfg.enabled) return;\n    if (cfg.promptSection) {\n      this.promptDispose = this.harness.systemPrompt.section({\n        name: ROW_ID,\n        order: SECTION_ORDER,\n        text: SECTION_TEXT.replaceAll(\"{PLANNER_MODEL}\", cfg.planner.model).replaceAll(\n          \"{EXECUTOR_MODEL}\",\n          cfg.executor.model,\n        ),\n      });\n    }\n    if (cfg.skill) {\n      this.skillDispose = this.harness.skills.register({\n        name: SKILL_NAME,\n        description: SKILL_DESCRIPTION,\n        whenToUse: SKILL_WHEN_TO_USE,\n        content: SKILL_CONTENT,\n        source: \"runtime\",\n      });\n    }\n  }\n}\n\nexport {\n  Config,\n  ModelRouter,\n  ModelRouter as default,\n  SETTINGS_NS,\n  name,\n  ROW_ID,\n  SKILL_CONTENT,\n  SKILL_DESCRIPTION,\n  SKILL_NAME,\n  SKILL_WHEN_TO_USE,\n  effortFor,\n  recentStepsHadError,\n  roleFor,\n  routeFor,\n};\nexport type {\n  AgentRole,\n  ModelRoute,\n  ReasoningEffort,\n  RoutingMode,\n  RouterConfig,\n} from \"./policy.js\";\n","/**\n * Pure routing policy for dsh-model-router: which model each agent role gets.\n * Kept free of Cordis imports so the policy is trivially unit-testable.\n * @module dsh-model-router/policy\n */\n\n/** Reasoning-effort levels a route may pin (mirrors the harness vocabulary). */\nexport type ReasoningEffort = \"off\" | \"low\" | \"high\" | \"max\";\n\n/**\n * How the router treats the root agent.\n * - `strict`: the root agent is always the planner (pro).\n * - `plan`: the root agent is pro only while plan mode is active; otherwise it\n *   falls back to the executor route, reserving pro for real planning.\n */\nexport type RoutingMode = \"strict\" | \"plan\";\n\n/** One route: a provider/model pair stamped onto an agent request. */\nexport interface ModelRoute {\n  provider: string;\n  model: string;\n  /**\n   * Optional reasoning-effort override. When omitted, the request inherits the\n   * session's own selection; when set, the router pins it for that role.\n   */\n  reasoningEffort?: ReasoningEffort;\n  /** Optional output-token cap for the role; omitted means inherit. */\n  maxTokens?: number;\n  /**\n   * Error-driven escalation (v1): when true, a failed execution step bumps the\n   * next request's effort to `escalateTo`, wearing off after `recoverySteps`\n   * clean steps. Deterministic, stateless — the session log is folded per\n   * request, so only *prior* steps are ever considered.\n   */\n  escalateOnError?: boolean;\n  /** Effort used for the request after a failed step. */\n  escalateTo?: ReasoningEffort;\n  /** Clean steps before escalation wears off. Defaults to 2. */\n  recoverySteps?: number;\n}\n\n/** The two roles the router distinguishes. */\nexport type AgentRole = \"planner\" | \"executor\";\n\n/** Resolved router configuration: one route per role plus routing mode. */\nexport interface RouterConfig {\n  planner: ModelRoute;\n  executor: ModelRoute;\n  mode: RoutingMode;\n  /**\n   * Live off-switch. Defaults to true; settable from Settings → Plugins →\n   * model-router (applies immediately) or from the patch row (next boot).\n   * When false, the router stops rewriting requests and unregisters the\n   * prompt section and the skill.\n   */\n  enabled: boolean;\n  promptSection: boolean;\n  skill: boolean;\n}\n\n/** Default recovery window: an error escalates for the next two completed steps. */\nexport const DEFAULT_RECOVERY_STEPS = 2;\n\n/**\n * Classify an agent as planner or executor.\n *\n * The main (root) agent of a session is the planner. Every agent created as a\n * delegation child — `subagent`, `subagent_fork`, workflow workers, ralph\n * rounds — is an executor. The harness stamps two durable facts on children:\n * `options.subagentDepth` (>= 1) and the session header `origin: \"subagent\"`.\n *\n * @param agent - the live agent (any subset of the runtime shape).\n * @returns the role the agent should be routed as.\n */\nexport function roleFor(agent: unknown): AgentRole {\n  const options = (agent as { options?: unknown })?.options;\n  const depth = (options as { subagentDepth?: unknown })?.subagentDepth;\n  if (typeof depth === \"number\" && depth >= 1) return \"executor\";\n  const session = (agent as { session?: unknown })?.session;\n  const origin = (session as { header?: unknown })?.header\n    ? ((session as { header: { origin?: unknown } }).header.origin)\n    : undefined;\n  if (origin === \"subagent\") return \"executor\";\n  return \"planner\";\n}\n\n/**\n * Resolve the route for one agent.\n * @param agent - the live agent.\n * @param config - the resolved router configuration.\n * @param planModeActive - whether plan mode is currently folded active for the\n *   agent's session; consulted only in `plan` routing mode.\n * @returns the model route to stamp, or `undefined` to leave the request alone.\n */\nexport function routeFor(\n  agent: unknown,\n  config: RouterConfig,\n  planModeActive = false,\n): ModelRoute | undefined {\n  const role = roleFor(agent);\n  if (role === \"executor\") return config.executor;\n  // Root agent. In `plan` mode, reserve the planner route for actual planning;\n  // otherwise the root falls back to the executor route.\n  if (config.mode === \"plan\" && !planModeActive) return config.executor;\n  return config.planner;\n}\n\n/**\n * Whether any of the last `recoverySteps` completed steps carried a failed\n * tool result. A failure is a `tool/result` event whose data carries an\n * `error` field (the harness records tool failures there).\n *\n * Steps are deduplicated by `turn:step`, and only *completed* steps count —\n * events are scanned from the tail, so the current in-flight request is never\n * considered.\n *\n * @param events - the agent's session event log (or `undefined`).\n * @param recoverySteps - how many completed steps back to scan.\n * @returns true when a failed step is within the window.\n */\nexport function recentStepsHadError(\n  events: readonly unknown[] | undefined,\n  recoverySteps: number = DEFAULT_RECOVERY_STEPS,\n): boolean {\n  if (!Array.isArray(events) || recoverySteps <= 0) return false;\n  const seen = new Set<string>();\n  let steps = 0;\n  for (let i = events.length - 1; i >= 0; i -= 1) {\n    const event = events[i] as\n      | { type?: string; data?: { turn?: number; step?: number; error?: unknown } }\n      | undefined;\n    if (event?.type !== \"tool/result\" || event.data === undefined) continue;\n    const key = `${event.data.turn}:${event.data.step}`;\n    if (!seen.has(key)) {\n      // A new step beyond the recovery window ends the scan; events for steps\n      // already inside the window are still checked below.\n      if (steps >= recoverySteps) break;\n      seen.add(key);\n      steps += 1;\n    }\n    if (event.data.error !== undefined && event.data.error !== null) return true;\n  }\n  return false;\n}\n\n/**\n * Resolve the reasoning effort to stamp for one request.\n *\n * Baseline is the route's `reasoningEffort`; when `escalateOnError` is enabled\n * and a recent step failed, the effort bumps to `escalateTo` (falling back to\n * the baseline when `escalateTo` is unset). Returns `undefined` to leave the\n * request's effort alone (inherit the session selection).\n *\n * @param route - the resolved route for the agent.\n * @param events - the agent's session event log.\n * @returns the effort to stamp, or `undefined` to inherit.\n */\nexport function effortFor(\n  route: ModelRoute,\n  events: readonly unknown[] | undefined,\n): ReasoningEffort | undefined {\n  if (route.escalateOnError === true && recentStepsHadError(events, route.recoverySteps)) {\n    return route.escalateTo ?? route.reasoningEffort;\n  }\n  return route.reasoningEffort;\n}\n"],"mappings":";AAoBA,SAAkB,eAAe;AACjC,OAAO,OAAO;AACd,SAAS,oBAAoB;AAC7B,SAAS,wBAAwB,yBAAyB;;;ACsCnD,IAAM,yBAAyB;AAa/B,SAAS,QAAQ,OAA2B;AACjD,QAAM,UAAW,OAAiC;AAClD,QAAM,QAAS,SAAyC;AACxD,MAAI,OAAO,UAAU,YAAY,SAAS,EAAG,QAAO;AACpD,QAAM,UAAW,OAAiC;AAClD,QAAM,SAAU,SAAkC,SAC5C,QAA6C,OAAO,SACtD;AACJ,MAAI,WAAW,WAAY,QAAO;AAClC,SAAO;AACT;AAUO,SAAS,SACd,OACA,QACA,iBAAiB,OACO;AACxB,QAAM,OAAO,QAAQ,KAAK;AAC1B,MAAI,SAAS,WAAY,QAAO,OAAO;AAGvC,MAAI,OAAO,SAAS,UAAU,CAAC,eAAgB,QAAO,OAAO;AAC7D,SAAO,OAAO;AAChB;AAeO,SAAS,oBACd,QACA,gBAAwB,wBACf;AACT,MAAI,CAAC,MAAM,QAAQ,MAAM,KAAK,iBAAiB,EAAG,QAAO;AACzD,QAAM,OAAO,oBAAI,IAAY;AAC7B,MAAI,QAAQ;AACZ,WAAS,IAAI,OAAO,SAAS,GAAG,KAAK,GAAG,KAAK,GAAG;AAC9C,UAAM,QAAQ,OAAO,CAAC;AAGtB,QAAI,OAAO,SAAS,iBAAiB,MAAM,SAAS,OAAW;AAC/D,UAAM,MAAM,GAAG,MAAM,KAAK,IAAI,IAAI,MAAM,KAAK,IAAI;AACjD,QAAI,CAAC,KAAK,IAAI,GAAG,GAAG;AAGlB,UAAI,SAAS,cAAe;AAC5B,WAAK,IAAI,GAAG;AACZ,eAAS;AAAA,IACX;AACA,QAAI,MAAM,KAAK,UAAU,UAAa,MAAM,KAAK,UAAU,KAAM,QAAO;AAAA,EAC1E;AACA,SAAO;AACT;AAcO,SAAS,UACd,OACA,QAC6B;AAC7B,MAAI,MAAM,oBAAoB,QAAQ,oBAAoB,QAAQ,MAAM,aAAa,GAAG;AACtF,WAAO,MAAM,cAAc,MAAM;AAAA,EACnC;AACA,SAAO,MAAM;AACf;;;AD1IA,IAAM,OAAO;AAGb,IAAM,mBAAmB,EAAE,OAAO;AAAA,EAChC,UAAU,EAAE,OAAO,EAAE,IAAI,CAAC;AAAA,EAC1B,OAAO,EAAE,OAAO,EAAE,IAAI,CAAC;AAAA,EACvB,iBAAiB,EAAE,MAAM,CAAC,OAAO,OAAO,QAAQ,KAAK,CAAC;AAAA,EACtD,WAAW,EAAE,OAAO,EAAE,IAAI,CAAC;AAAA,EAC3B,iBAAiB,EAAE,QAAQ;AAAA,EAC3B,YAAY,EAAE,MAAM,CAAC,OAAO,OAAO,QAAQ,KAAK,CAAC;AAAA,EACjD,eAAe,EAAE,OAAO,EAAE,IAAI,CAAC;AACjC,CAAC;AAGD,IAAM,SAAS,EAAE,OAAO;AAAA,EACtB,SAAS,iBAAiB,QAAQ;AAAA,IAChC,UAAU;AAAA,IACV,OAAO;AAAA,EACT,CAAU;AAAA,EACV,UAAU,iBAAiB,QAAQ;AAAA,IACjC,UAAU;AAAA,IACV,OAAO;AAAA,EACT,CAAU;AAAA,EACV,MAAM,EAAE,MAAM,CAAC,UAAU,MAAM,CAAC,EAAE,QAAQ,QAAQ;AAAA,EAClD,SAAS,EAAE,QAAQ,EAAE,QAAQ,IAAI;AAAA,EACjC,eAAe,EAAE,QAAQ,EAAE,QAAQ,IAAI;AAAA,EACvC,OAAO,EAAE,QAAQ,EAAE,QAAQ,IAAI;AACjC,CAAC;AAGD,IAAM,cAAc,kBAAkB,cAAc;AAQpD,SAAS,cAAc,KAA4B;AACjD,QAAM,SAAS,OAAO,OAAO,CAAC,CAAC;AAC/B,SAAO;AAAA,IACL,SAAS,OAAO;AAAA,IAChB,UAAU,OAAO;AAAA,IACjB,MAAM,OAAO;AAAA,IACb,SAAS,OAAO;AAAA,IAChB,eAAe,OAAO;AAAA,IACtB,OAAO,OAAO;AAAA,EAChB;AACF;AAMA,IAAM,gBAAgB;AAEtB,IAAM,eAAe;AAErB,IAAM,aAAa;AAEnB,IAAM,oBACJ;AAEF,IAAM,oBAAoB;AAE1B,IAAM,gBAAgB;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AA+BtB,IAAM,SAAS;AA2Cf,SAAS,iBAAiB,OAA2B;AACnD,QAAM,SAAS,MAAM,SAAS;AAC9B,MAAI,CAAC,MAAM,QAAQ,MAAM,EAAG,QAAO;AACnC,MAAI;AACF,WAAO,aAAa,MAA4C;AAAA,EAClE,QAAQ;AACN,WAAO;AAAA,EACT;AACF;AAKA,IAAM,cAAN,cAA0B,QAAQ;AAAA,EAChC,OAAO,SAAS,CAAC,UAAU,cAAc;AAAA,EAEzC;AAAA;AAAA,EAGA;AAAA;AAAA,EAGA;AAAA;AAAA,EAGA;AAAA;AAAA,EAGA;AAAA,EAEA,YAAY,KAAc,YAAqB,CAAC,GAAG;AACjD,UAAM,KAAK,aAAa;AACxB,SAAK,SAAS,cAAc,SAAS;AACrC,SAAK,SAAS,MAAM,KAAK;AACzB,SAAK,UAAU;AAIf,SAAK,QAAQ,GAAG,iBAAiB,CAAC,EAAE,MAAM,MAAM;AAM9C,YAAM,UAAU,MAAM,IAAI;AAAA,QACxB;AAAA,QACA,OAAO,SAAS,SAAS;AACvB,gBAAM,WAAW,MAAM,KAAK;AAC5B,gBAAM,MAAM,KAAK,OAAO;AACxB,cAAI,CAAC,IAAI,QAAS,QAAO;AACzB,gBAAM,QAAQ,SAAS,OAAO,KAAK,iBAAiB,KAAK,CAAC;AAC1D,cAAI,UAAU,OAAW,QAAO;AAChC,gBAAM,UAAmC;AAAA,YACvC,GAAG;AAAA,YACH,UAAU,MAAM;AAAA,YAChB,OAAO,MAAM;AAAA,UACf;AACA,cAAI,MAAM,cAAc,OAAW,SAAQ,YAAY,MAAM;AAC7D,gBAAM,SAAS,UAAU,OAAO,MAAM,SAAS,MAAM;AACrD,cAAI,WAAW,OAAW,SAAQ,kBAAkB;AACpD,iBAAO;AAAA,QACT;AAAA,QACA,EAAE,SAAS,KAAK;AAAA,MAClB;AACA,WAAK,QAAQ,GAAG,kBAAkB,CAAC,aAAa;AAC9C,YAAI,aAAa,MAAO,SAAQ;AAAA,MAClC,CAAC;AAAA,IACH,CAAC;AAID,SAAK,OAAO,KAAK,MAAM;AACvB;AAAA,MACE;AAAA,MACA;AAAA,MACA;AAAA,MACA,KAAK;AAAA,MACL;AAAA,QACE,WAAW,CAAC,YAAY;AACtB,eAAK,SAAS;AAAA,QAChB;AAAA,QACA,UAAU,MAAM,KAAK,OAAO,KAAK,OAAO,CAAC;AAAA,MAC3C;AAAA,IACF;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAQQ,OAAO,KAAyB;AACtC,QAAI,KAAK,eAAe;AACtB,WAAK,cAAc;AACnB,WAAK,gBAAgB;AAAA,IACvB;AACA,QAAI,KAAK,cAAc;AACrB,WAAK,aAAa;AAClB,WAAK,eAAe;AAAA,IACtB;AACA,QAAI,CAAC,IAAI,QAAS;AAClB,QAAI,IAAI,eAAe;AACrB,WAAK,gBAAgB,KAAK,QAAQ,aAAa,QAAQ;AAAA,QACrD,MAAM;AAAA,QACN,OAAO;AAAA,QACP,MAAM,aAAa,WAAW,mBAAmB,IAAI,QAAQ,KAAK,EAAE;AAAA,UAClE;AAAA,UACA,IAAI,SAAS;AAAA,QACf;AAAA,MACF,CAAC;AAAA,IACH;AACA,QAAI,IAAI,OAAO;AACb,WAAK,eAAe,KAAK,QAAQ,OAAO,SAAS;AAAA,QAC/C,MAAM;AAAA,QACN,aAAa;AAAA,QACb,WAAW;AAAA,QACX,SAAS;AAAA,QACT,QAAQ;AAAA,MACV,CAAC;AAAA,IACH;AAAA,EACF;AACF;","names":[]}

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "dsh-model-router",
   "description": "DeepSeek Harness plugin: role-based model routing — the planner agent runs on deepseek-v4-pro, delegated executor subagents run on deepseek-v4-flash.",
-  "version": "0.2.0",
+  "version": "0.4.0",
   "publishConfig": {
     "access": "public"
   },
@@ -35,12 +35,14 @@
   "peerDependencies": {
     "@deepseek-ai/cordis": "^4.0.1",
     "@deepseek-ai/schemastery": "^3.18.1",
-    "@deepseek-ai/dsh-plan-mode": "^0.1.0-rc.8"
+    "@deepseek-ai/dsh-plan-mode": "^0.1.0-rc.8",
+    "@deepseek-ai/dsh-settings": "^0.1.0-rc.8"
   },
   "devDependencies": {
     "@deepseek-ai/cordis": "^4.0.1",
     "@deepseek-ai/schemastery": "^3.18.1",
     "@deepseek-ai/dsh-plan-mode": "^0.1.0-rc.8",
+    "@deepseek-ai/dsh-settings": "^0.1.0-rc.8",
     "@types/node": "^22.10.0",
     "tsup": "^8.3.5",
     "typescript": "^5.7.2",