ultimate-pi 0.12.0 → 0.13.0

package/.agents/skills/ck-search/SKILL.md

@@ -1,99 +1,23 @@
 ---
 name: ck-search
-description: "Semantic code search using ck. Use when exploring codebases, finding related code, or searching by concept rather than exact text. Triggers on: search for, find code that, where is, look for patterns, find all files related to, find similar code, explore codebase, semantic search."
+description: "DEPRECATED — ck-search was removed from ultimate-pi. Use cocoindex-search or /skill:ccc instead. Triggers retained for backward compatibility: ck, ck-search, semantic search."
 ---
 
-# ck-search: Semantic Code Search
+# ck-search (deprecated)
 
-## When to Use
+**`ck` / `@beaconbay/ck-search` is no longer used in this harness.**
 
-Use `ck` instead of `grep`/`find` for **any codebase exploration**. Reserve raw grep for only: exact literal string match (specific error message, exact function name).
+Use instead:
 
-## Quick Reference
+- **`/skill:cocoindex-search`** or **`/skill:ccc`** — CocoIndex Code (`ccc search`)
+- **graphify** — architecture, callers, communities (`graphify query`, `explain`, `path`)
 
-```bash
-# Hybrid (best default — lexical + semantic fused via RRF)
-ck --hybrid "query" .
-
-# Semantic only (conceptual — finds by meaning)
-ck --sem "concept" src/
-
-# Lexical BM25 (phrase-based, better than grep for multi-word)
-ck --lex "phrase" .
-
-# Grep-compatible (exact match, same flags as grep)
-ck "exact string" -rn src/
-```
-
-## Search Decision Tree
-
-```
-Agent needs to find code
-  ├─ Exact literal string (error msg, function name) → grep/rg
-  ├─ Conceptual / multi-word → ck --hybrid
-  ├─ Find similar patterns → ck --sem
-  └─ Unsure → ck --hybrid (safe default)
-```
-
-## Key Flags
-
-| Flag | Purpose | When |
-|------|---------|------|
-| `--hybrid` | BM25 + semantic RRF fusion | **Default for exploration** |
-| `--sem` | Semantic only (embedding similarity) | Conceptual: "error handling", "auth flow" |
-| `--lex` | BM25 lexical only | Phrase search without regex |
-| `--limit N` | Top N results | Keep output lean (default 10) |
-| `--threshold 0.7` | Min similarity score | Filter low-confidence results |
-| `--json` | Machine-readable output | When piping to other tools |
-| `-n` | Line numbers | Same as grep |
-| `-C N` | Context lines | Same as grep |
-| `-r` | Recursive | Same as grep |
-| `-l` | Files with matches | List matching files only |
-
-## Index Management
-
-```bash
-ck --status .              # Check if index exists
-ck index .                 # Build/rebuild full index
-ck --add file.ts           # Add single file to index
-ck --clean .               # Remove index (rebuild from scratch)
-ck --switch-model MODEL    # Rebuild with different embedding model
-```
-
-## Usage Patterns
+## Migration
 
-### Find related code
 ```bash
-ck --hybrid "retry logic with exponential backoff" .
+uv tool install 'cocoindex-code[full]'
+bash "$UP_PKG/.pi/scripts/harness-cocoindex-bootstrap.sh"
+ccc search --limit 10 "your query"
 ```
 
-### Find authentication code
-```bash
-ck --sem "authentication middleware" src/ --limit 20
-```
-
-### Find error handling patterns
-```bash
-ck --hybrid "error handling and recovery" src/ --limit 15
-```
-
-### Find database code
-```bash
-ck --sem "database connection pool" src/
-```
-
-### Exact function name (still use grep)
-```bash
-grep -rn "processPayment" src/
-```
-
-## Integration Notes
-
-- **Index location**: `.ck/index/` in project root (gitignored)
-- **First search auto-indexes**: `ck --sem` builds index on first run if missing
-- **Fully offline**: No API keys, no network, embeddings run locally
-- **MCP mode**: `ck --serve` exposes ck_search/ck_get/ck_info/ck_reindex as MCP tools (future integration)
-
-## Token Efficiency
-
-ck results are ranked and scored. Use `--limit` to cap output. A typical ck --hybrid call returns 10 results (~500-1000 tokens) vs raw grep which can return hundreds of unranked matches (~5000-20000 tokens).
+Remove legacy index: `rm -rf .ck` (optional).

package/.agents/skills/cocoindex-search/SKILL.md

@@ -0,0 +1,35 @@
+---
+name: cocoindex-search
+description: "Semantic code search using CocoIndex Code (ccc). Use when exploring codebases, finding related implementation by meaning, or replacing legacy ck-search. Triggers on: semantic code search, ccc, cocoindex, cocoindex-code, find code related to, search the codebase for implementation, /skill:ck-search."
+---
+
+# cocoindex-search
+
+CocoIndex Code (`ccc`) provides offline, AST-aware semantic search over the project codebase.
+
+## Quick start
+
+```bash
+ccc search --limit 10 "harness subagent policy"
+ccc status
+```
+
+## Full reference
+
+Load the vendored skill: **`/skill:ccc`** (`.pi/skills/ccc/SKILL.md`).
+
+## Harness lanes
+
+| Question type | Tool |
+|---------------|------|
+| Callers, callees, cross-module paths | `graphify explain` / `graphify path` |
+| Implementation by meaning | `ccc search --limit N "…"` |
+| Structural patterns | `sg -p '…'` |
+
+## Setup
+
+```bash
+bash "$UP_PKG/.pi/scripts/harness-cocoindex-bootstrap.sh"
+```
+
+Indexing before harness scouts is automatic — do not run `ccc index` or `ccc search --refresh` in `scout-semantic`.

package/.agents/skills/harness-orchestration/SKILL.md

@@ -36,7 +36,7 @@ LIMIT 30
 1. **Parallel `tasks`** — one `subagent({ tasks: [...] })` for scouts, decompose+hypothesis, or review fan-in; subprocesses run in parallel upstream.
 2. **Blocking calls** — each `subagent` returns when the subprocess exits; no `get_subagent_result` polling.
 3. **Compact handoffs** — pass scout/decompose JSON only; never paste full subprocess message logs into the next spawn.
-4. **Spawn caps** — bridge enforces **8** active + **12** total harness spawns per session (`PI_SUBAGENT_TIMEOUT_MS` / per-task `timeoutMs` for backstop).
+4. **Spawn caps** — bridge enforces **8** active + **12** total harness spawns per session. Do **not** pass `timeoutMs` unless the user wants a cap — subprocesses wait for natural exit (`PI_SUBAGENT_TIMEOUT_MS` optional env backstop only).
 
 ## Command → agent
 
@@ -71,9 +71,9 @@ Spawn `harness/evaluator` / `harness/adversary` via `subagent` in the **same** p
 {
   "agentScope": "both",
   "tasks": [
-    { "agent": "harness/planning/scout-graphify", "task": "…", "timeoutMs": 90000 },
-    { "agent": "harness/planning/scout-structure", "task": "…", "timeoutMs": 90000 },
-    { "agent": "harness/planning/scout-semantic", "task": "…", "timeoutMs": 90000 }
+    { "agent": "harness/planning/scout-graphify", "task": "…" },
+    { "agent": "harness/planning/scout-structure", "task": "…" },
+    { "agent": "harness/planning/scout-semantic", "task": "…" }
   ]
 }
 ```

package/.pi/PACKAGING.md

@@ -24,6 +24,7 @@ We use an explicit allowlist (not the whole `.pi/` tree) so dev-only artifacts n
 - Ship `.pi/settings.example.json`, not `.pi/settings.json` (dev checkout uses `".."` local package)
 - Include **`vendor/pi-model-router/`** ([`pi-model-router`](https://github.com/yeliu84/pi-model-router), MIT) — see repo [`THIRD_PARTY_NOTICES.md`](../THIRD_PARTY_NOTICES.md); refresh with `npm run vendor:sync-router`
 - Include **`vendor/pi-vcc/`** ([`pi-vcc`](https://github.com/sting8k/pi-vcc), MIT; inspired by [lllyasviel/VCC](https://github.com/lllyasviel/VCC)) — loaded via `.pi/extensions/ultimate-pi-vcc.ts`; refresh with `npm run vendor:sync-vcc`
+- Include **`vendor/pi-subagents/`** (vendored from [narumiruna/pi-extensions](https://github.com/narumiruna/pi-extensions) `pi-subagents`) — loaded via `.pi/extensions/harness-subagents.ts`; refresh with `npm run vendor:sync-subagents`
 
 ## Settings
 

package/.pi/SYSTEM.md

@@ -81,41 +81,42 @@ edges at build time. Use these to answer call-graph questions without external t
 - **How does `Auth` reach `Database`?** → `graphify path "Auth" "Database"` (shortest call chain)
 - **Trace a dependency chain deep** → `graphify query "how does X depend on Y" --dfs`
 
-**Semantic code search via graphify:**
-Graphify already indexes the entire codebase as a knowledge graph. Use graphify
-for conceptual code search before falling back to `ck`:
-- **Find code by meaning** → `graphify query "where is authentication logic"`
-- **Find related concepts** → `graphify query "what connects to error handling"`
+**Semantic code search (two lanes):**
+- **Architecture / relationships** → graphify (`query`, `explain`, `path`, `GRAPH_REPORT.md`)
+- **Implementation by meaning** → CocoIndex Code (`ccc search --limit N "concept"`)
+
+Examples:
+- **Find code by meaning** → `ccc search --limit 10 "authentication session validation"`
+- **Who calls X / cross-module path** → `graphify explain "X"` or `graphify path "A" "B"`
 - **Cross-file surprises** → `graphify query "what unexpected connections exist"`
 
 **Order of operations for codebase exploration:**
 1. Read `graphify-out/GRAPH_REPORT.md` (god nodes, surprises, suggested questions)
-2. Run `graphify query` for domain-specific questions, call traces, and semantic search
-3. Use `graphify explain "Concept"` for caller/callee/dependency deep dives
-4. Use `sg -p 'pattern'` for structural code search, then `ck --hybrid` only if graph and ast-grep don't surface it
-5. Read individual files last — the graph already told you what matters
+2. Run `graphify query` / `explain` / `path` for architecture and call graphs
+3. Use `sg -p 'pattern'` for structural code search
+4. Use `ccc search --limit N` for conceptual implementation chunks when graphify/sg are insufficient
+5. Read individual files last — scouts and graph already narrowed the set
+
+**Indexing:** Harness runs incremental `ccc index` before subagent spawns. Use `ccc search` only in agents; run `ccc index` at session start or after large edits on parent turns. Never use `ccc search --refresh` in scouts. `/skill:ccc` for full CLI reference.
 
 ### Fallback Search (when graph doesn't cover it)
 
-> [!note] Graphify handles semantic search and call graphs
-> Graphify already provides semantic code search and call-graph tracing. Use
-> `graphify query`, `graphify explain`, and `graphify path` as your primary
-> code exploration tools. Only fall back to `sg`/`ck`/`find` when the graph
-> doesn't have the answer (e.g., not yet indexed, or you need exact raw text).
+> [!note] Graphify + ccc split responsibilities
+> Graphify owns call graphs and cross-module relationships. `ccc` owns AST-aware
+> semantic chunks. Only fall back to `find`/`grep` for exact literals or non-code files.
 
 | Tool | When | Command |
 |------|------|---------|
-| `sg -p` | **Primary code search** — AST-aware structural pattern matching | `sg -p 'pattern' --lang typescript` |
+| `sg -p` | **Structural code search** — AST pattern matching | `sg -p 'pattern' --lang typescript` |
 | `sg scan` | Rule-based code scanning (use project rules in `sgconfig.yml`) | `sg scan` |
-| `ck --hybrid` | Lexical + semantic fusion search (fallback after ast-grep) | `ck --hybrid "query" .` |
-| `ck --sem` | Purely conceptual searches (fallback after ast-grep) | `ck --sem "concept" src/` |
+| `ccc search` | **Semantic chunks** — implementation by meaning | `ccc search --limit 10 "query"` |
 | `find` | File discovery by name/glob only | `find . -name "*.ts"` |
 | `grep` | **Last resort** — exact literal string matching in non-code files only | `grep -F "exact string"` |
 
-- **Always prefer ast-grep (`sg`) over grep for code search.** ast-grep understands code structure via tree-sitter — it matches patterns, not strings. Use it for: finding function calls, class definitions, import statements, variable usage, and any structural code query.
+- **Always prefer ast-grep (`sg`) over grep for code search.** ast-grep understands code structure via tree-sitter — it matches patterns, not strings.
 - Never use grep for code search. grep is only for: log files, non-code text files, exact byte-level matching when AST patterns can't work.
-- Always use `--limit N` on ck to cap output and save context.
-- Graphify is primary. ast-grep is secondary. ck/find are fallbacks. grep is last resort.
+- Always use `--limit N` on `ccc search` to cap output and save context.
+- Graphify is primary for architecture. ast-grep is secondary for structure. ccc is semantic implementation search. grep is last resort.
 - Do NOT install or use grepai/seagoat/mgrep for call-graph traces or semantic
   search — graphify already handles both.
 

package/.pi/agents/harness/planning/scout-graphify.md

@@ -15,6 +15,8 @@ Explore the codebase via graphify for the task in `HarnessSpawnContext`. You do
 
 Findings should feed **constraints, prior art, and tensions** for the decompose agent (existing patterns, god nodes, surprising connections).
 
+**Lane contract:** you own **relationships and architecture** (`graphify query`, `explain`, `path`). `scout-semantic` owns implementation-by-meaning via `ccc search` — do not duplicate semantic chunk search here.
+
 ## Spawn context
 
 Read `HarnessSpawnContext` in the spawn prompt (`task_summary`, `mode`, `plan_packet_path`, `risk_level`, `quick`). For `mode: revise`, read the existing plan at `plan_packet_path` first and focus findings on what changed or is at risk.

package/.pi/agents/harness/planning/scout-semantic.md

@@ -1,5 +1,5 @@
 ---
-description: Plan-phase scout — ck semantic code search (read-only).
+description: Plan-phase scout — CocoIndex semantic code search (read-only).
 tools: read, bash, ls
 disallowed_tools: write, edit, ask_user, approve_plan, create_plan, subagent, grep, find
 extensions: false
@@ -11,7 +11,9 @@ You are the **Harness planning scout (semantic lane)**.
 
 ## Mission
 
-Find conceptually related code via ck semantic search for the task in `HarnessSpawnContext`. You do **not** build the PlanPacket or mutate files.
+Find conceptually related **implementation** via CocoIndex (`ccc search`) for the task in `HarnessSpawnContext`. You do **not** build the PlanPacket or mutate files.
+
+**Lane contract:** `scout-graphify` owns relationships, callers, and communities. You own **meaning** — functions, classes, and chunks that implement the task.
 
 ## Spawn context
 
@@ -19,13 +21,18 @@ Read `HarnessSpawnContext` in the spawn prompt. For `mode: revise`, bias searche
 
 ## Process
 
-1. Use `ck search` or `ck query` (or project-documented ck CLI) with task-focused queries.
-2. If ck is unavailable, set `status: partial` and document in `findings`.
-3. **Stop early** — top **5** most relevant paths only.
+1. Run **2–3** task-focused queries: `ccc search "<query>" --limit 5` (add `--path` when spawn context names a directory).
+2. The harness runs incremental `ccc index` before scouts spawn — **do not** run `ccc index`, `ccc init`, or `ccc search --refresh`.
+3. If `ccc` is missing or the index is empty: `status: partial` and document in `findings`.
+4. **Stop early** — top **5** most relevant paths only.
 
 ## Bash guardrails
 
-Read-only only: no installs, index rebuilds that mutate disk, or redirects.
+Read-only only: no installs, indexing, daemon control, or redirects.
+
+**Allowed:** `ccc search`, `ccc status`, `ls`, `head`, `cat`, `sed -n` (read slices).
+
+**Forbidden:** `ccc index`, `ccc init`, `ccc reset`, `ccc daemon`, `ccc search --refresh`, package installs.
 
 ## Output limits
 

package/.pi/extensions/harness-run-context.ts

@@ -55,6 +55,10 @@ import {
 	parseStructuredDocument,
 	writeYamlFile,
 } from "../lib/harness-yaml.js";
+import { claimExtensionLoad } from "./lib/extension-load-guard.js";
+
+// @ts-expect-error pi extensions run as ESM
+const MODULE_URL = import.meta.url;
 
 interface SessionEntryLike {
 	type?: string;
@@ -191,6 +195,7 @@ function needsClarificationFollowUp(ctx: HarnessRunContext | null): boolean {
 }
 
 export default function harnessRunContext(pi: ExtensionAPI) {
+	if (!claimExtensionLoad("harness-run-context", MODULE_URL)) return;
 	let activeCtx: HarnessRunContext | null = null;
 
 	pi.on("session_start", async (_event, ctx) => {

package/.pi/extensions/harness-subagents.ts

@@ -1,14 +1,25 @@
 /**
  * harness-subagents — vendored pi-subagents with ultimate-pi discovery and policy gates.
+ *
+ * Dynamic-imports the bridge only after claimExtensionLoad so a stale global npm
+ * install (missing vendor/pi-subagents) does not crash local development in this repo.
  */
 
+import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 import { claimExtensionLoad } from "./lib/extension-load-guard.js";
-import { getHarnessPackageRoot } from "./lib/harness-paths.js";
-import { createHarnessSubagentsExtension } from "./lib/harness-subagents-bridge.js";
 
 // @ts-expect-error pi extensions run as ESM
 const MODULE_URL = import.meta.url;
 
-export default claimExtensionLoad("harness-subagents", MODULE_URL)
-	? createHarnessSubagentsExtension(getHarnessPackageRoot(MODULE_URL))
-	: () => {};
+async function loadHarnessSubagents(): Promise<(pi: ExtensionAPI) => void> {
+	if (!claimExtensionLoad("harness-subagents", MODULE_URL)) {
+		return () => {};
+	}
+	const { getHarnessPackageRoot } = await import("./lib/harness-paths.js");
+	const { createHarnessSubagentsExtension } = await import(
+		"./lib/harness-subagents-bridge.js"
+	);
+	return createHarnessSubagentsExtension(getHarnessPackageRoot(MODULE_URL));
+}
+
+export default await loadHarnessSubagents();

package/.pi/extensions/lib/harness-cocoindex-refresh.ts

@@ -0,0 +1,49 @@
+/**
+ * Incremental CocoIndex refresh before harness subagent batches (plan/execute).
+ * Agents use `ccc search` only; harness owns `ccc index`.
+ */
+
+import { spawnSync } from "node:child_process";
+import { existsSync } from "node:fs";
+import { join } from "node:path";
+
+const DEFAULT_TIMEOUT_MS = 120_000;
+
+export function refreshHarnessCocoindexIndex(cwd: string): string | undefined {
+	if (process.env.HARNESS_COCOINDEX_REFRESH === "0") {
+		return undefined;
+	}
+	const settingsPath = join(cwd, ".cocoindex_code", "settings.yml");
+	if (!existsSync(settingsPath)) {
+		return undefined;
+	}
+
+	const timeoutMs = Number(
+		process.env.HARNESS_COCOINDEX_REFRESH_TIMEOUT_MS ?? DEFAULT_TIMEOUT_MS,
+	);
+	const result = spawnSync("ccc", ["index"], {
+		cwd,
+		encoding: "utf8",
+		timeout: Number.isFinite(timeoutMs) ? timeoutMs : DEFAULT_TIMEOUT_MS,
+		stdio: "pipe",
+	});
+
+	if (result.error) {
+		const msg = `harness-cocoindex: ccc index failed (${result.error.message})`;
+		if (process.env.HARNESS_COCOINDEX_REFRESH_STRICT === "1") {
+			return msg;
+		}
+		return `${msg} — continuing`;
+	}
+
+	if (result.status !== 0) {
+		const stderr = (result.stderr ?? "").trim().slice(0, 500);
+		const msg = `harness-cocoindex: ccc index exited ${result.status ?? "?"}${stderr ? `: ${stderr}` : ""}`;
+		if (process.env.HARNESS_COCOINDEX_REFRESH_STRICT === "1") {
+			return msg;
+		}
+		return `${msg} — continuing`;
+	}
+
+	return undefined;
+}

package/.pi/extensions/lib/harness-subagent-policy.ts

@@ -24,9 +24,13 @@ const PLANNING_BASH_DENY_PATTERNS = [
 	/\bgraphify\s+update\b/i,
 	/\bgraphify\s+extract\b/i,
 	/\bgraphify\s+install\b/i,
+	/\bccc\s+(index|init|reset|daemon)\b/i,
+	/\bccc\s+search\b.*--refresh/i,
 	/\bpip\s+install\b/i,
 	/\buv\s+tool\s+install\b/i,
 	/\bnpm\s+install\b/i,
+	/\bnpm\s+install\b.*cocoindex/i,
+	/\buv\s+tool\s+install\b.*cocoindex/i,
 ];
 
 const BASH_MUTATION_PATTERNS = [
@@ -141,7 +145,7 @@ export function evaluateHarnessSubagentToolCall(
 			return {
 				action: "block",
 				reason:
-					"harness-subagent-policy: planning scouts may use read-only graphify/sg/ck commands only.",
+					"harness-subagent-policy: planning scouts may use read-only graphify/sg/ccc commands only.",
 			};
 		}
 	}

package/.pi/extensions/lib/harness-subagents-bridge.ts

@@ -12,6 +12,7 @@ import {
 	type HarnessSubagentsOptions,
 	type SpawnAuthForward,
 } from "../../../vendor/pi-subagents/src/subagents.js";
+import { refreshHarnessCocoindexIndex } from "./harness-cocoindex-refresh.js";
 import { captureHarnessEvent } from "./harness-posthog.js";
 import {
 	checkHarnessSpawnBudget,
@@ -32,37 +33,6 @@ import {
 const spawnBudget = createSpawnBudgetState();
 let lastSessionId = "harness";
 
-function maskApiKey(key: string | undefined): string | undefined {
-	if (!key) return undefined;
-	if (key.length <= 12) return "***";
-	return `${key.slice(0, 7)}…${key.slice(-4)}`;
-}
-
-// #region agent log
-function agentDebugLog(
-	hypothesisId: string,
-	location: string,
-	message: string,
-	data: Record<string, unknown>,
-): void {
-	fetch("http://127.0.0.1:7928/ingest/a5d40896-34cb-4f12-97db-df7ada0b22f0", {
-		method: "POST",
-		headers: {
-			"Content-Type": "application/json",
-			"X-Debug-Session-Id": "e762d5",
-		},
-		body: JSON.stringify({
-			sessionId: "e762d5",
-			hypothesisId,
-			location,
-			message,
-			data,
-			timestamp: Date.now(),
-		}),
-	}).catch(() => {});
-}
-// #endregion
-
 async function resolveHarnessSpawnAuth(
 	ctx: ExtensionContext,
 	agent: AgentConfig,
@@ -72,43 +42,11 @@ async function resolveHarnessSpawnAuth(
 		: undefined;
 	const concrete = resolveConcreteSubagentModel(ctx.cwd, parentModel, agent);
 	if (!concrete) {
-		// #region agent log
-		agentDebugLog(
-			"D",
-			"harness-subagents-bridge.ts:resolveHarnessSpawnAuth",
-			"no concrete model",
-			{
-				agent: agent.name,
-				agentModel: agent.model,
-				parentModel: parentModel
-					? `${parentModel.provider}/${parentModel.id}`
-					: undefined,
-			},
-		);
-		// #endregion
 		return undefined;
 	}
 	const apiKey = await ctx.modelRegistry.getApiKeyForProvider(
 		concrete.provider,
 	);
-	// #region agent log
-	agentDebugLog(
-		"F",
-		"harness-subagents-bridge.ts:resolveHarnessSpawnAuth",
-		"concrete subprocess auth",
-		{
-			agent: agent.name,
-			parentModel: parentModel
-				? `${parentModel.provider}/${parentModel.id}`
-				: undefined,
-			concreteModel: concrete.modelRef,
-			routerProfile: concrete.routerProfile,
-			routerTier: concrete.routerTier,
-			apiKey: maskApiKey(apiKey),
-			usable: isUsableApiKey(apiKey),
-		},
-	);
-	// #endregion
 	if (!isUsableApiKey(apiKey)) return undefined;
 	return {
 		provider: concrete.provider,
@@ -145,6 +83,14 @@ export function createHarnessSubagentsExtension(
 				if (!pre.ok) {
 					return { ok: false, message: pre.message };
 				}
+				if (phase === "plan" || phase === "execute") {
+					const refreshMsg = refreshHarnessCocoindexIndex(ctx.cwd);
+					if (refreshMsg?.includes("continuing")) {
+						// warn-only path; do not block spawn
+					} else if (refreshMsg) {
+						return { ok: false, message: refreshMsg };
+					}
+				}
 			}
 			return { ok: true };
 		},

package/.pi/harness/docs/adrs/0033-parent-orchestrated-planning.md

@@ -19,7 +19,7 @@
 ### Positive
 
 - Full plan visible in parent session; editor `plan-review.md` path surfaced from parent.
-- Specialized scouts (graphify, ast-grep, ck) run in parallel with clear JSON contracts.
+- Specialized scouts (graphify, ast-grep, cocoindex `ccc`) run in parallel with clear JSON contracts.
 - Pre-approval adversary separate from post-run `harness/adversary`.
 
 ### Negative

package/.pi/prompts/harness-plan.md

@@ -30,8 +30,9 @@ Read **harness-debate-plan** skill before Review Gate rounds.
 
 1. Use `subagent` with `agentScope: "both"` and parallel `tasks` where lanes are independent.
 2. Each `subagent` call blocks until subprocesses finish — batch parallel scouts in one `tasks` array.
-3. Cap: **12** harness subagent invocations per parent session (extension-enforced).
-4. Compact task text: embed `HarnessSpawnContext` JSON + lane-specific instructions only.
+3. Do **not** set `timeoutMs` unless the user explicitly requests a cap — subagents run until natural completion (optional backstop: `PI_SUBAGENT_TIMEOUT_MS`).
+4. Cap: **12** harness subagent invocations per parent session (extension-enforced).
+5. Compact task text: embed `HarnessSpawnContext` JSON + lane-specific instructions only.
 
 ## Step 0 — Parse `$ARGUMENTS`
 
@@ -44,19 +45,23 @@ Read **harness-debate-plan** skill before Review Gate rounds.
 
 Use `[HarnessActivePlan]` / `[HarnessRunContext]` only. On revise: preserve `plan_id` / `task_id`. Canonical paths: `plan-packet.yaml`, `research-brief.yaml`, `artifacts/*.yaml`.
 
+## Phase 0 — Semantic index (automatic)
+
+Do **not** run `ccc index` or `ccc search --refresh`. The harness runs incremental `ccc index` before subagent spawns. Proceed directly to Phase 1 scouts.
+
 ## Phase 1 — Parallel scouts
 
 ```json
 {
   "agentScope": "both",
   "tasks": [
-    { "agent": "harness/planning/scout-graphify", "task": "<HarnessSpawnContext + graphify lane>", "timeoutMs": 90000 },
-    { "agent": "harness/planning/scout-structure", "task": "<HarnessSpawnContext + structure lane>", "timeoutMs": 90000 }
+    { "agent": "harness/planning/scout-graphify", "task": "<HarnessSpawnContext + graphify lane>" },
+    { "agent": "harness/planning/scout-structure", "task": "<HarnessSpawnContext + structure lane>" }
   ]
 }
 ```
 
-Add `harness/planning/scout-semantic` to `tasks` unless `--quick`. Require graphify + structure success.
+Add `harness/planning/scout-semantic` to `tasks` unless `--quick`. Require graphify + structure success. Semantic lane uses `ccc search` only (see `scout-semantic` agent).
 
 ## Phase 2 & 3 — Decompose + hypothesis (parallel)
 

package/.pi/prompts/harness-setup.md

@@ -156,7 +156,7 @@ bash "$UP_PKG/.pi/scripts/harness-cli-verify.sh"
 
 **Required (script must exit 0):** scrapling + harness-web smoke, ctx7, biome, ast-grep (`sg`), sentrux (when harness manifest present).
 
-**Warnings allowed:** gh (if not authenticated), agent-browser (if OS libs need manual `sudo apt-get install`), ck (empty corpus on tiny repos).
+**Warnings allowed:** gh (if not authenticated), agent-browser (if OS libs need manual `sudo apt-get install`), cocoindex-code (empty corpus on tiny repos; first `[full]` install downloads local embedding model).
 
 If the script reports **agent-browser shared library errors** on Linux/WSL, run the fix it prints, then re-verify:
 
@@ -232,20 +232,24 @@ if [ ! -f .pi/harness/browser.json ]; then
 fi
 ```
 
-### 2.4 — ck-search (Semantic Code Search)
+### 2.4 — CocoIndex Code (semantic code search)
+
+Offline semantic search via [CocoIndex Code](https://cocoindex.io/cocoindex-code/) (`ccc`). **CLI only** — do not register `ccc mcp` in harness setup (MCP defaults to refresh-before-search and is slower).
 
 ```bash
-if ! command -v ck &>/dev/null || [ "$FORCE" = "true" ]; then
-	npm install -g @beaconbay/ck-search
-fi
+command -v uv &>/dev/null || curl -LsSf https://astral.sh/uv/install.sh | sh
+export PATH="$HOME/.local/bin:$PATH"
+uv tool install 'cocoindex-code[full]'   # or: pipx install 'cocoindex-code[full]'
+bash "$UP_PKG/.pi/scripts/harness-cocoindex-bootstrap.sh"
+# Rebuild index: bash "$UP_PKG/.pi/scripts/harness-cocoindex-bootstrap.sh" --force
+ccc status
 ```
 
-Verify: `ck --version`
+**Indexing before scouts is automatic:** `harness-subagents` runs incremental `ccc index` before plan/execute subagent batches. Agents use `ccc search` only.
 
-Register as MCP server (if Claude MCP available):
-```bash
-claude mcp list 2>/dev/null && claude mcp add ck-search -s user -- ck --serve || echo "MCP not available — ck will be used as CLI only"
-```
+**First install:** `[full]` pulls sentence-transformers + torch (~hundreds of MB–1GB+). Run bootstrap once; do not run `ccc index` inside scouts.
+
+Verify: `ccc doctor` and `ccc search --limit 3 "export function"` (no `--refresh`).
 
 ### 2.5 — biome (Lint + Format Gate)
 
@@ -666,7 +670,7 @@ Output summary table:
 | scrapling / harness-web | ✓/✗ | Auth: yes/no |
 | ctx7 | ✓/✗ | Login: yes/no |
 | agent-browser | ✓/✗ | Config: .pi/harness/browser.json |
-| ck-search | ✓/✗ | MCP: registered/CLI-only |
+| cocoindex-code | ✓/✗ | `ccc status`; index auto-refreshed before harness scouts |
 | biome | ✓/✗ | Project config: found/default |
 | ast-grep | ✓/✗ | AST-aware code search (`sg`)
 | gh CLI | ✓/✗ | Auth: yes/no |

package/.pi/scripts/README.md

@@ -21,6 +21,7 @@ From **Typescript extensions**, use `resolveHarnessScript()` / `getHarnessPackag
 | Action | Command |
 |--------|---------|
 | Graphify bootstrap | `bash "$UP_PKG/.pi/scripts/harness-graphify-bootstrap.sh"` |
+| CocoIndex Code bootstrap | `bash "$UP_PKG/.pi/scripts/harness-cocoindex-bootstrap.sh"` |
 | CLI tool install + smoke tests | `bash "$UP_PKG/.pi/scripts/harness-cli-verify.sh"` |
 | Deterministic harness checks | `node "$UP_PKG/.pi/scripts/harness-verify.mjs"` |
 | Sentrux rules bootstrap (harness-setup) | `node "$UP_PKG/.pi/scripts/harness-sentrux-bootstrap.mjs"` |