@kylebrodeur/pi-model-router 0.2.0 → 0.3.0

package/CHANGELOG.md

@@ -1,3 +1,8 @@
+## [0.2.1] - 2026-06-05
+
+- Bumped @earendil-works/pi-coding-agent SDK to 0.78.1.
+- Updated tool `execute` signatures to match the new SDK API.
+
 # Changelog
 
 All notable changes to this project will be documented in this file.
@@ -7,6 +12,16 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.3.0] - 2026-06-28
+
+### Removed
+- Removed router-owned Ollama sync and the models.json writer; Ollama registration is delegated to @kylebrodeur/pi-model-discovery.
+- Removed pi-qmd-ledger integration config and checked-in ledger JSONL artifacts.
+- Removed checked-in pi-session HTML artifacts and ignored future pi-session-*.html files.
+
+### Changed
+- /router ollama-sync now delegates to /providers sync when pi-model-discovery is installed and otherwise shows install guidance.
+
 ## [0.2.0] - 2026-05-17
 
 ### Added
@@ -50,7 +65,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Scope shim module
 - Progressive enhancement with plugin detection
 
-[Unreleased]: https://github.com/kylebrodeur/pi-model-router/compare/v0.2.0...HEAD
+[Unreleased]: https://github.com/kylebrodeur/pi-model-router/compare/v0.3.0...HEAD
+[0.3.0]: https://github.com/kylebrodeur/pi-model-router/compare/v0.2.2...v0.3.0
 [0.2.0]: https://github.com/kylebrodeur/pi-model-router/compare/v0.1.4...v0.2.0
 [0.1.4]: https://github.com/kylebrodeur/pi-model-router/compare/v0.1.3...v0.1.4
 [0.1.3]: https://github.com/kylebrodeur/pi-model-router/compare/v0.1.2...v0.1.3

package/CONTRIBUTING.md

@@ -10,7 +10,7 @@ yeliu84/pi-model-router (upstream)
     └── your-github/pi-model-router-fork (this repo)
             │
             ├── branch: main (tracks upstream)
-            ├── branch: feature/ollama-sync
+            ├── branch: feature/model-discovery-shim
             ├── branch: feature/rate-limit-fallback
             └── branch: feature/feature-toggles
 ```
@@ -23,7 +23,7 @@ Separate your work into **independent PRs** for maximum merge probability:
 **Commit message:** `feat: add feature toggle system for modular extensions`
 
 **Files touched:**
-- `extensions/types.ts` — add `FeatureToggles`, `OllamaSyncConfig`, `RateLimitFallbackConfig`
+- `extensions/types.ts` — add `FeatureToggles` and `RateLimitFallbackConfig`
 - `extensions/config.ts` — add feature fields to `mergeConfig()` and `normalizeConfig()`
 
 **Rationale:**
@@ -32,17 +32,18 @@ Separate your work into **independent PRs** for maximum merge probability:
 - Enables future features
 - Easy to review and merge
 
-### PR 2: Ollama Auto-Sync Feature
-**Commit message:** `feat: add ollama-sync for auto-registering local models`
+### PR 2: Model Discovery Compatibility Shim
+**Commit message:** `feat: delegate ollama discovery to pi-model-discovery`
 
 **Files touched:**
-- `extensions/ollama-sync.ts` — new file
-- `extensions/index.ts` — conditionally initialize from toggles
+- `extensions/model-discovery.ts` — discovery command detection helper
+- `extensions/commands.ts` — compatibility shim for `/router ollama-sync`
+- `extensions/index.ts` — advisory warning when model discovery is absent
 
 **Rationale:**
-- Self-contained module
-- Uses existing patterns (pi.exec, config merge)
-- Useful to upstream community (many Pi users use Ollama)
+- Keeps existing user command muscle memory
+- Avoids duplicating provider discovery ownership
+- Leaves local model registration to pi-model-discovery
 
 ### PR 3: Rate Limit Fallback
 **Commit message:** `feat: add rate-limit fallback with manual restore`
@@ -134,15 +135,13 @@ For each PR:
 Follow upstream conventions:
 
 ```
-feat: add ollama-sync for auto-registering local models
+feat: delegate ollama discovery to pi-model-discovery
 
-- Detect new Ollama models on session start/reload
-- Infer capabilities (vision, reasoning, context window)
-- Update models.json with sensible defaults
-- Toggle via features.ollamaSync in config
+- Detect the pi-model-discovery `/providers` command when available
+- Delegate `/router ollama-sync` to `/providers sync`
+- Show install guidance when model discovery is absent
 
 Refs: #upstream-issue-if-any
-```
 
 ```
 feat: add rate-limit fallback to local Ollama models
@@ -159,7 +158,7 @@ Refs: #upstream-issue-if-any
 feat: add feature toggle system for modular extension
 
 - features config key for enabling/disabling features
-- ollamaSync and rateLimitFallback as first features
+- rateLimitFallback as the first feature
 - Per-turn routing preserved as default behavior
 - Project-level override support
 ```
@@ -270,14 +269,14 @@ Your options:
 ### PR 1: Feature Toggles (minimal)
 | File | Change |
 |------|--------|
-| `extensions/types.ts` | +FeatureToggles, +OllamaSyncConfig, +RateLimitFallbackConfig, add fields to RouterConfig |
+| `extensions/types.ts` | +FeatureToggles, +RateLimitFallbackConfig, add fields to RouterConfig |
 | `extensions/config.ts` | preserve feature fields in mergeConfig/normalizeConfig |
 
-### PR 2: Ollama Sync
+### PR 2: Model Discovery Compatibility Shim
 | File | Change |
 |------|--------|
-| `extensions/ollama-sync.ts` | New file |
-| `extensions/index.ts` | Import + conditional init |
+| `extensions/model-discovery.ts` | New helper file |
+| `extensions/commands.ts` | Delegate `/router ollama-sync` |
 
 ### PR 3: Rate Limit Fallback
 | File | Change |

package/LEARNINGS.md

@@ -20,7 +20,7 @@ The extension strictly separates responsibilities into dedicated modules:
 *   `extensions/state.ts`: Session-persisted state management
 *   `extensions/ui.ts`: Status line and widget rendering
 *   `extensions/commands.ts`: CLI command registrations and completions
-*   `extensions/feature-module.ts`: Feature-specific modules (e.g., `ollama-sync.ts`, `rate-limit.ts`)
+*   `extensions/feature-module.ts`: Feature-specific modules (e.g., `rate-limit.ts`)
 
 *   **When to use**: For any extension with complex logic that will benefit from maintainability and testing benefits.
 *   **Key benefit**: Each module can be unit tested in isolation. Changes to one concern (e.g., updating heuristics) don't risk breaking unrelated logic (e.g., command registration).
@@ -86,7 +86,6 @@ A `features` object in `model-router.json` allows enabling/disabling features at
 ```json
 {
   "features": {
-    "ollamaSync": true,
     "respectPiScope": false,
     "rateLimitFallback": true
   }
@@ -175,7 +174,7 @@ Commands are registered under `/router` with subcommands (`/router status`, `/ro
     -   Add a "Tier Selection" guide, explaining when to use `high` vs. `medium` vs. `low`.
 
 3.  **Future Feature Ideas**
-    -   Per-profile Ollama models (different models per router profile).
+    -   Per-profile local models registered by pi-model-discovery.
     -   Rate limit dashboard (`/router rate-limit show` with history graph).
     -   Config UI wizard (`/router config setup` interactive).
 

package/QUICKSTART.md

@@ -22,29 +22,25 @@ In Pi, run:
 /reload
 ```
 
-You should see in console:
-
-```
-[router] Feature sync complete - ollama: true rate-limit: true
-[router] ollama-sync: enabled
-[router] rate-limit-fallback: enabled
-```
+You should see the router initialize and report feature sync without an Ollama sync feature state.
 
 ## 4. Verify Features
 
-### Auto-Sync Test
+### Model Discovery Note
+
+Install `pi-model-discovery` when you want local Ollama models registered with Pi:
 
 ```bash
-# In another terminal
-ollama pull llama3.2:3b
+pi install npm:@kylebrodeur/pi-model-discovery
 ```
 
+Then run:
+
 ```
-# In Pi
-/router ollama-sync
+/providers sync
 ```
 
-Expected: `Added: llama3.2:3b`
+Use `/router ollama-sync` only as a delegation shortcut to `/providers sync`.
 
 ### Fallback Test
 
@@ -56,7 +52,7 @@ Expected: `Added: llama3.2:3b`
 /router fallback
 ```
 
-Expected: switches to best available model matching your `fallbackSequence` (e.g., an Ollama model)
+Expected: switches to best available model matching your `fallbackSequence` (for example, an Ollama model registered by pi-model-discovery)
 
 ```
 # Restore
@@ -67,7 +63,7 @@ Expected: switches to best available model matching your `fallbackSequence` (e.g
 
 | Command               | What it does                               |
 | --------------------- | ------------------------------------------ |
-| `/router ollama-sync` | Manually sync Ollama models                |
+| `/router ollama-sync` | Delegate to /providers sync                |
 | `/router fallback`    | Switch to fallback model sequence (manual) |
 | `/router restore`     | Restore original cloud model               |
 | `/router init`        | Scaffold default config file               |
@@ -80,31 +76,28 @@ All existing `/router *` commands still work unchanged.
 ## 6. Daily Workflow
 
 ```bash
-# Pull new Ollama model
+# Pull a new Ollama model when using local fallback models
 ollama pull qwen3.5:7b
 ```
 
 ```
-# In Pi — auto-sync triggers on session start
-/new
-# → [Router] Added 1 model(s): qwen3.5:7b
-# → Run /reload to see: qwen3.5:7b
+# In Pi — model registration belongs to pi-model-discovery
+/providers sync
 
-/reload
-# → Model now available in /model
+# Optional router compatibility shortcut
+/router ollama-sync
 ```
 
 ## Troubleshooting
 
-**"Ollama not available"**
-→ Check `ollama list` works in terminal
+**Ollama models missing**
+→ Install/run pi-model-discovery and check `/providers status`
 
 **Models not in registry**
-→ `/reload` — extension reload refreshes model registry
+→ `/providers sync` from pi-model-discovery, then `/reload` if the Pi UI needs a refresh
 
 **TypeScript errors**
 → Requires Pi 0.67+ for rate-limit fallback (uses `after_provider_response` event)
-→ Ollama sync works with current Pi version
 
 **Config not applying**
 → Verify `~/.pi/agent/model-router.json` exists

package/README.md

@@ -4,7 +4,7 @@
 [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)
 [![Pi Version](https://img.shields.io/badge/pi-%3E%3D0.70.2-blue)](https://github.com/mariozechner/pi-coding-agent)
 
-Intelligent per-turn model router extension for the [pi](https://github.com/mariozechner/pi-coding-agent) coding agent. Automatically selects between high, medium, and low-tier LLMs based on task intent, session budget, context size, and custom rules — with automatic fallbacks, phase awareness, Ollama sync, and transparent rate-limit recovery.
+Intelligent per-turn model router extension for the [pi](https://github.com/mariozechner/pi-coding-agent) coding agent. Automatically selects between high, medium, and low-tier LLMs based on task intent, session budget, context size, and custom rules — with automatic fallbacks, phase awareness, and transparent rate-limit recovery.
 
 ## ✨ Features
 
@@ -21,10 +21,9 @@ Intelligent per-turn model router extension for the [pi](https://github.com/mari
 | **Phase Memory** | Biased stickiness to keep you in the same tier during multi-turn planning or implementation work. |
 | **Thinking Control** | Full control over reasoning/thinking levels per tier and profile. |
 | **Persistent State** | Pins, profiles, costs, and debug history are remembered across agent restarts and conversation branches. |
-| **Ollama Auto-Sync** | Auto-detect and register local Ollama models. |
 | **Rate-Limit Fallback** | Detect HTTP 402/429/503/529 and transparently fall back to local models. |
 | **Feature Toggles** | Enable/disable features at user or project level. |
-| **Progressive Enhancement** | Detects installed plugins (qmd-ledger, agent-bus) and integrates conditionally. |
+| **Optional Agent Bus Integration** | Detects agent-bus when installed and integrates conditionally. |
 
 ## 📦 Installation
 
@@ -105,7 +104,6 @@ Copy the example config to one of:
 ```json
 {
   "features": {
-    "ollamaSync": true,
     "rateLimitFallback": true,
     "scopeShim": true,
     "perTurnRouting": true,
@@ -113,7 +111,6 @@ Copy the example config to one of:
     "costBudgeting": true,
     "phaseMemory": true,
     "contextCompression": true,
-    "ledgerIntegration": false,
     "agentBusIntegration": false
   }
 }
@@ -121,6 +118,16 @@ Copy the example config to one of:
 
 **Priority:** Project config `.pi/model-router.json` overrides user config `~/.pi/agent/model-router.json`. Both override defaults.
 
+### Model Discovery
+
+Install `@kylebrodeur/pi-model-discovery` with:
+
+```bash
+pi install npm:@kylebrodeur/pi-model-discovery
+```
+
+Run `/providers sync` to discover Ollama models. `/router ollama-sync` only delegates to `/providers sync`. The router no longer writes `~/.pi/agent/models.json`.
+
 ### Rate Limit Interception & Fallback
 
 The router can gracefully handle 429 Rate Limit and Quota errors. If the error specifies a wait time (e.g., "reset after 58s"), the router will pause and automatically retry the prompt if the wait time is under your threshold. If it exceeds the threshold or is unparseable, it fails over to the next available model in your fallback sequence.
@@ -151,9 +158,8 @@ After installing optional extensions, copy one of these to `.pi/model-router.jso
 
 | File | Feature |
 |------|---------|
-| `model-router.ledger.json` | Log routing decisions to qmd-ledger |
 | `model-router.agent-bus.json` | Publish model changes to agent-bus |
-| `model-router.essential.json` | All integrations enabled |
+| `model-router.essential.json` | Rate-limit fallback, scope sync, and agent-bus integration |
 
 ## ⌨️ Commands
 
@@ -177,7 +183,7 @@ After installing optional extensions, copy one of these to `.pi/model-router.jso
 
 | Command | Feature | Description |
 |---------|---------|-------------|
-| `/router ollama-sync` | ollamaSync | Manually sync Ollama models |
+| `/router ollama-sync` | | Delegates to /providers sync from pi-model-discovery |
 | `/router fallback` | rateLimitFallback | Switch to fallback model sequence |
 | `/router restore` | rateLimitFallback | Restore cloud model |
 | `/router init` | | Scaffold default config file |
@@ -188,7 +194,7 @@ After installing optional extensions, copy one of these to `.pi/model-router.jso
 ## 🔒 Requirements
 
 - **Pi >= 0.70.2** — Required for `after_provider_response` event (rate limit detection)
-- **Ollama** — Running locally for Ollama sync and fallback
+- **Ollama** — Optional and only needed for local fallback models; install pi-model-discovery for Ollama model registration.
 - **Node.js >= 18** — For TypeScript compilation
 
 ## 📚 Documentation

package/TESTING.md

@@ -4,8 +4,7 @@
 
 - [ ] **Pi updated** to 0.67+ (for `after_provider_response` event support)
 - [ ] **Node.js** 18+ available
-- [ ] **Ollama installed and running**
-- [ ] At least one Ollama model pulled locally
+- [ ] **Ollama plus pi-model-discovery are required only for local fallback/delegation checks**
 
 ## Installation Steps
 
@@ -41,6 +40,7 @@ pi
 
 - [ ] Shows current profile, tier, model
 - [ ] Shows `Router enabled: true`
+- [ ] `Features:` line includes `rateLimit=` and does not include the removed Ollama sync state
 
 ### Test Routing Decision
 
@@ -66,18 +66,11 @@ Design a full-stack app architecture with React, Node, and PostgreSQL
 
 - [ ] Request uses high-tier model
 
-## Test 2: Ollama Auto-Sync
+## Test 2: Model Discovery Delegation
 
-**Goal:** Verify new Ollama models are detected and added.
+**Goal:** Verify `/router ollama-sync` delegates Ollama discovery ownership to pi-model-discovery.
 
-### Step 1: Pull a New Model
-
-```bash
-# In another terminal
-ollama pull llama3.2:3b
-```
-
-### Step 2: Trigger Sync in Pi
+### Check 1: Without pi-model-discovery installed
 
 ```
 # In Pi TUI
@@ -86,49 +79,25 @@ ollama pull llama3.2:3b
 
 **Expected:**
 
-- [ ] Notification: `[Router] Added 1 model(s)`
-- [ ] Notification: `Run /reload to see: llama3.2:3b`
-
-### Step 3: Reload and Verify
-
-```
-/reload
-```
-
-**Expected:**
-
-- [ ] Extension reloads
-- [ ] New model appears in `/model` selector under "ollama"
-
-### Step 4: Auto-Sync on Session Start
-
-```
-/model  # Switch to a non-router model
-/new     # Start new session
-```
+- [ ] Notification: `Install pi-model-discovery for Ollama model discovery: pi install npm:@kylebrodeur/pi-model-discovery. Then run /providers sync.`
+- [ ] Router does not access `models.json`
+- [ ] Router does not execute `ollama`
 
-**Expected:**
-
-- [ ] On session start, Ollama sync runs automatically
-- [ ] If new models exist, notifications appear
-- [ ] Console log: `[router] ollama-sync: feature enabled`
-
-### Step 5: Project-Level Config Override
+### Check 2: With pi-model-discovery installed
 
 ```bash
-# In your project directory
-echo '{ "features": { "ollamaSync": false } }' > .pi/model-router.json
+pi install npm:@kylebrodeur/pi-model-discovery
 ```
 
 ```
-# In Pi TUI (from project directory)
-/router
+# In Pi TUI
+/router ollama-sync
 ```
 
 **Expected:**
 
-- [ ] Auto-sync doesn't run on session start
-- [ ] Manual `/router ollama-sync` still works
+- [ ] First notification: `Delegating Ollama model discovery to /providers sync.`
+- [ ] Provider extension handles `/providers sync`
 
 ## Test 3: Rate Limit Fallback
 
@@ -197,7 +166,6 @@ echo '{ "features": { "rateLimitFallback": false } }' > ~/.pi/agent/model-router
 cat > ~/.pi/agent/model-router.json << 'EOF'
 {
   "features": {
-    "ollamaSync": true,
     "rateLimitFallback": true,
     "perTurnRouting": true
   }
@@ -212,7 +180,6 @@ mkdir -p .pi
 cat > .pi/model-router.json << 'EOF'
 {
   "features": {
-    "ollamaSync": false,
     "rateLimitFallback": false
   }
 }
@@ -229,31 +196,11 @@ EOF
 **Expected:**
 
 - [ ] Console shows `[router] rate-limit-fallback: disabled`
-- [ ] Console shows `[router] ollama-sync: disabled`
 - [ ] Per-turn routing still works (project didn't override it)
 
 ## Test 5: Edge Cases
 
-### Missing Ollama
-
-```
-# Stop Ollama
-ollama stop  # or kill process
-
-# Try sync
-/router ollama-sync
-```
-
-**Expected:**
-
-- [ ] Notification: `Ollama not available`
-- [ ] No crash, graceful failure
-
-### Missing models.json
-
-```bash
-mv ~/.pi/agent/models.json ~/.pi/agent/models.json.bak
-```
+### Missing Model Discovery Extension
 
 ```
 /router ollama-sync
@@ -261,18 +208,10 @@ mv ~/.pi/agent/models.json ~/.pi/agent/models.json.bak
 
 **Expected:**
 
-- [ ] Notification: `models.json not found`
-
-```bash
-# Restore
-mv ~/.pi/agent/models.json.bak ~/.pi/agent/models.json
-```
+- [ ] Notification: `Install pi-model-discovery for Ollama model discovery: pi install npm:@kylebrodeur/pi-model-discovery. Then run /providers sync.`
+- [ ] No crash, graceful guidance
 
-### No Ollama Models Configured
-
-```bash
-# Temporarily move models.json ollama section
-```
+### No Fallback Models Configured
 
 ```
 /router fallback
@@ -280,16 +219,12 @@ mv ~/.pi/agent/models.json.bak ~/.pi/agent/models.json
 
 **Expected:**
 
-- [ ] `No Ollama models available` error notification
+- [ ] `No fallback models available` or equivalent error notification
+- [ ] Router remains on the current model
 
 ## Test 6: Combined Features (Full Workflow)
 
-**Goal:** Verify all features work together.
-
-```bash
-# Pull new model
-ollama pull qwen3.5:7b
-```
+**Goal:** Verify router features work together.
 
 ```
 # Pi session
@@ -299,13 +234,12 @@ ollama pull qwen3.5:7b
 **Expected:**
 
 1. [ ] Session starts
-2. [ ] Auto-sync detects `qwen3.5:7b`
-3. [ ] Notification: `Added 1 model(s)`
-4. [ ] `/reload` prompt shown
-5. [ ] After `/reload`, routing profiles work
-6. [ ] `/router profile balanced` switches to router
-7. [ ] Simple query routes to low tier (possibly Ollama)
-8. [ ] Complex query routes to high tier (cloud)
+2. [ ] `/router status` shows no removed Ollama sync feature state
+3. [ ] `/router config` includes `rateLimitFallback` and does not list the removed Ollama sync feature
+4. [ ] `/router config ollama-sync` reports `Unknown feature: ollama-sync. Try: rate-limit, classifier, budget, respect-pi-scope`
+5. [ ] `/router profile balanced` switches to router
+6. [ ] Simple query routes to low tier
+7. [ ] Complex query routes to high tier (cloud)
 
 ## Regression Tests (Upstream Feature Checklist)
 
@@ -331,7 +265,7 @@ ollama pull qwen3.5:7b
 ```bash
 # Check file paths
 ls -la ~/.pi/agent/extensions/pi-model-router/
-# Should see: index.ts, config.ts, types.ts, routing.ts, provider.ts, commands.ts, state.ts, ui.ts, constants.ts, ollama-sync.ts, rate-limit.ts, features.ts
+# Should see: index.ts, config.ts, types.ts, routing.ts, provider.ts, commands.ts, state.ts, ui.ts, constants.ts, model-discovery.ts, rate-limit.ts
 ```
 
 ### TypeScript Errors After Update
@@ -363,12 +297,13 @@ npm install -g @mariozechner/pi-coding-agent
 ### Fallback Model Not In Registry
 
 ```
-/reload
-/model  # Check if new models appear
+/providers status
+/providers sync
+/model  # Check if local models appear
 ```
 
-If Ollama models don't appear after `/reload`:
+If Ollama models don't appear after `/providers sync`:
 
-1. Check `models.json` has ollama provider section
-2. Verify `baseUrl` is correct: `http://127.0.0.1:11434/v1`
-3. Check Ollama is running: `ollama list` in terminal
+1. Verify pi-model-discovery is installed: `/providers status`
+2. Check Ollama is running: `ollama list` in terminal
+3. Pull at least one local model for fallback use

package/docs/pi-model-router-refactor-prompt.md

@@ -0,0 +1,99 @@
+# Agent Prompt: Remove ollama-sync overlap from pi-model-router
+
+## Context
+
+`@kylebrodeur/pi-model-router` contains an `extensions/ollama-sync.ts` module that auto-discovers and registers Ollama models. This functionality has been superseded by `@kylebrodeur/pi-model-discovery` which does the same thing better:
+
+| Feature | `ollama-sync.ts` (router) | `pi-model-discovery` |
+|---------|--------------------------|---------------------|
+| Discovery API | Runs `ollama list` + keyword heuristics | Native `/api/tags` + `/api/show` |
+| Registration | Writes to `models.json` (legacy format) | `pi.registerProvider()` (no reload needed) |
+| Timing | `session_start` hook | Async factory (available immediately) |
+| Caching | None | TTL cache at `ollama-model-cache.json` |
+| Scope | Manually updates `models.json` | Writes `settings.json` `enabledModels` |
+| Commands | `/router ollama-sync` | `/providers sync`, `/providers status`, etc. |
+
+## Task
+
+Refactor `@kylebrodeur/pi-model-router` to remove the duplicated Ollama discovery logic and instead delegate to `pi-model-discovery` when it's installed.
+
+## Repo
+
+`https://github.com/kylebrodeur/pi-model-router` — clone and work from main.
+
+## Changes Required
+
+### 1. Remove `extensions/ollama-sync.ts` internal implementation
+
+The `performOllamaSync()` and `initializeOllamaSync()` functions should be removed. The `/router ollama-sync` command and its backing logic should either:
+- (A) Check if `pi-model-discovery` is installed and delegate to `/providers sync` if so
+- (B) Be removed entirely with a notice that `pi-model-discovery` handles this
+
+Recommendation: (A) — check for discovery extension, delegate or show helpful message.
+
+### 2. Update `extensions/index.ts`
+
+Remove the import and initialization of `initializeOllamaSync`. Replace with a detection check:
+
+```ts
+// Check if pi-model-discovery is installed; if not, warn user
+const hasModelDiscovery = /* check pi extension registry */;
+if (!hasModelDiscovery) {
+  ctx.ui.notify(
+    "Router: Install pi-model-discovery for Ollama auto-sync (pi install npm:@kylebrodeur/pi-model-discovery)",
+    "warning"
+  );
+}
+```
+
+### 3. Update `/router ollama-sync` command in `extensions/commands.ts`
+
+Instead of running the internal sync, either:
+- Delegate: `pi.sendUserMessage("/providers sync")` if pi-model-discovery is installed
+- Or remove the command and redirect users to `/providers sync`
+
+### 4. Update `extensions/config.ts`
+
+Remove the `OllamaSyncConfig` type and related config from the router config schema. Remove `features.ollamaSync` toggle — it's no longer the router's concern.
+
+### 5. Update `model-router.example.json` and related config files
+
+Remove `features.ollamaSync` entries. Add a comment pointing to `pi-model-discovery`.
+
+### 6. Update README
+
+- Remove the ollama-sync section from Features table
+- Add a "Model Discovery" section explaining that `pi-model-discovery` handles this
+- Update the `/router ollama-sync` command entry to note it delegates to `/providers sync`
+
+### 7. Update CHANGELOG
+
+Document the removal as a "replaced by pi-model-discovery" note.
+
+## What to Keep
+
+- `extensions/rate-limit.ts` — this is NOT in pi-model-discovery (rate-limit fallback is router-specific)
+- `extensions/routing.ts` — core routing logic, stays
+- `extensions/provider.ts` — router provider registration, stays
+- All other router features (cost budgeting, phase memory, intent classifier, etc.)
+
+## Detection pattern for pi extensions
+
+To check if `pi-model-discovery` is installed, look for:
+```ts
+// Check if the extension is registered
+const hasDiscovery = pi.getRegisteredExtensions?.()
+  ?.some(e => e.name?.includes('pi-model-discovery')) ?? false;
+```
+
+Or check if the `/providers` command is available via the command registry.
+
+## Acceptance Criteria
+
+- `extensions/ollama-sync.ts` is deleted
+- `features.ollamaSync` config key removed
+- `/router ollama-sync` command either removed or delegates to `/providers sync`
+- Router still works without pi-model-discovery (graceful degradation)
+- README updated
+- Tests pass (`pnpm test`)
+- No breaking changes to rate-limit, routing, or cost budgeting features