ultimate-pi 0.16.0 → 0.18.0

package/.pi/scripts/harness-model-router-routing.test.mjs

@@ -0,0 +1,97 @@
+#!/usr/bin/env node
+/**
+ * Unit tests for session-locked pi-model-router routing (no LLM).
+ * Run: npx tsx .pi/scripts/harness-model-router-routing.test.mjs
+ */
+
+import assert from "node:assert/strict";
+import { readFileSync } from "node:fs";
+import { join, dirname } from "node:path";
+import { fileURLToPath } from "node:url";
+import {
+	decideSessionLock,
+	applyThinkingToDecision,
+	buildRoutingDecision,
+	decideRouting,
+} from "../../vendor/pi-model-router/extensions/routing.js";
+
+const ROOT = join(dirname(fileURLToPath(import.meta.url)), "..", "..");
+
+const sampleProfile = {
+	high: { model: "openai/gpt-5.5", thinking: "high" },
+	medium: { model: "openai/gpt-5.5", thinking: "medium" },
+	low: { model: "openai/gpt-5.5", thinking: "low" },
+};
+
+const planningContext = {
+	systemPrompt: "You are a harness architect. Design tradeoffs and migration strategy.",
+	messages: [
+		{
+			role: "user",
+			content:
+				"Plan a multi-phase refactor across modules with architecture review.",
+			timestamp: 1,
+		},
+	],
+};
+
+const shortContext = {
+	systemPrompt: "Summarize briefly.",
+	messages: [{ role: "user", content: "changelog", timestamp: 1 }],
+};
+
+const lockHigh = decideSessionLock(
+	planningContext,
+	"auto",
+	sampleProfile,
+	undefined,
+	undefined,
+	0.5,
+	[{ matches: "changelog", tier: "low" }],
+);
+assert.equal(lockHigh.tier, "high", "planning prompt locks high tier");
+
+const lockLow = decideSessionLock(shortContext, "auto", sampleProfile);
+assert.equal(lockLow.tier, "low", "short summary locks low tier");
+
+const locked = buildRoutingDecision(
+	"auto",
+	sampleProfile,
+	lockHigh.tier,
+	"planning",
+	lockHigh.reasoning,
+);
+const thinkingTurn = decideRouting(
+	{
+		...planningContext,
+		messages: [
+			...planningContext.messages,
+			{ role: "user", content: "changelog only", timestamp: 2 },
+		],
+	},
+	"auto",
+	sampleProfile,
+	locked,
+);
+const merged = applyThinkingToDecision(locked, thinkingTurn, sampleProfile);
+assert.equal(merged.targetLabel, locked.targetLabel, "model stays locked");
+assert.equal(merged.tier, thinkingTurn.tier, "thinking tier follows turn");
+assert.equal(merged.thinking, "low", "low thinking from turn tier config");
+
+const examplePath = join(ROOT, ".pi", "model-router.example.json");
+const example = JSON.parse(readFileSync(examplePath, "utf8"));
+for (const [name, profile] of Object.entries(example.profiles ?? {})) {
+	const { high, medium, low } = profile;
+	assert.equal(
+		high.model,
+		medium.model,
+		`example profile ${name}: medium/high same model`,
+	);
+	assert.equal(
+		medium.model,
+		low.model,
+		`example profile ${name}: low/medium same model`,
+	);
+}
+
+console.log("harness-model-router-routing.test: PASS");

package/.pi/scripts/harness-sync-model-router.mjs

@@ -29,11 +29,24 @@ function saveSettings(settingsPath, data) {
 	);
 }
 
+function readDefaultRouterProfile(configPath) {
+	if (!existsSync(configPath)) return "auto";
+	try {
+		const data = JSON.parse(readFileSync(configPath, "utf8"));
+		const profile =
+			typeof data.defaultProfile === "string" ? data.defaultProfile.trim() : "";
+		return profile || "auto";
+	} catch {
+		return "auto";
+	}
+}
+
 function main() {
 	const root = process.cwd();
 	const configPath = join(root, ".pi", "model-router.json");
 	const settingsPath = join(root, ".pi", "settings.json");
 	const hasConfig = existsSync(configPath);
+	const defaultRouterProfile = readDefaultRouterProfile(configPath);
 
 	const settings = loadSettings(settingsPath);
 	if (!settings) {
@@ -67,14 +80,14 @@ function main() {
 
 	if (noProjectDefault) {
 		settings.defaultProvider = "router";
-		settings.defaultModel = "auto";
+		settings.defaultModel = defaultRouterProfile;
 		changed = true;
 	}
 
 	if (changed) {
 		saveSettings(settingsPath, settings);
 		console.log(
-			"✓ Router defaults set (`router` / `auto`) — run /reload in pi when ready",
+			`✓ Router defaults set (\`router\` / \`${defaultRouterProfile}\`) — run /reload in pi when ready`,
 		);
 	} else {
 		console.log("[harness-model-router] Defaults unchanged (user set defaultProvider)");

package/.pi/scripts/harness-verify.mjs

@@ -23,6 +23,8 @@ const REQUIRED_SCHEMAS = [
 	"eval-verdict.schema.json",
 	"harness-spawn-context.schema.json",
 	"harness-turn.schema.json",
+	"sentrux-manifest-proposal.schema.json",
+	"sentrux-signal.schema.json",
 ];
 
 const REQUIRED_ADRS = [
@@ -39,6 +41,7 @@ const REQUIRED_ADRS = [
 	"0032-harness-command-orchestration.md",
 	"0037-subagent-submit-tools.md",
 	"0038-budget-telemetry-only.md",
+	"0040-practice-grounded-orchestration.md",
 ];
 
 const REQUIRED_EXTENSIONS = [
@@ -145,6 +148,34 @@ async function checkSentruxRules() {
 	ok(".sentrux/rules.toml present");
 }
 
+async function checkModelRouterThinkingOnly() {
+	const path = join(ROOT, ".pi", "model-router.json");
+	if (!(await fileExists(path))) {
+		ok("model-router.json absent (skip thinking-only tier check)");
+		return;
+	}
+	let raw;
+	try {
+		raw = JSON.parse(await readFile(path, "utf-8"));
+	} catch {
+		fail("invalid .pi/model-router.json");
+	}
+	const profiles = raw.profiles ?? {};
+	for (const [name, profile] of Object.entries(profiles)) {
+		const high = profile?.high?.model;
+		const medium = profile?.medium?.model;
+		const low = profile?.low?.model;
+		if (
+			!(high && medium && low && high === medium && medium === low)
+		) {
+			fail(
+				`model-router profile "${name}" must use the same model on high/medium/low (thinking-only tiers)`,
+			);
+		}
+	}
+	ok("model-router.json thinking-only (same model per profile)");
+}
+
 async function checkSentruxGate() {
 	await checkSentruxRules();
 
@@ -152,13 +183,21 @@ async function checkSentruxGate() {
 		ok("Sentrux MCP stub gate skipped (HARNESS_SENTRUX_REQUIRED not set)");
 		return;
 	}
-	const stubPath = join(ROOT, ".pi", "harness", "evals", "smoke", "sentrux-stub.json");
-	if (!(await fileExists(stubPath))) {
-		fail(
-			"HARNESS_SENTRUX_REQUIRED=true but .pi/harness/evals/smoke/sentrux-stub.json missing",
-		);
+	const runDir = process.env.HARNESS_RUN_DIR?.trim();
+	const runSignalPath = runDir
+		? join(runDir, "artifacts", "sentrux-signal.yaml")
+		: null;
+	if (runSignalPath && (await fileExists(runSignalPath))) {
+		ok(`Sentrux run signal present (${runSignalPath})`);
+	} else {
+		const stubPath = join(ROOT, ".pi", "harness", "evals", "smoke", "sentrux-stub.json");
+		if (!(await fileExists(stubPath))) {
+			fail(
+				"HARNESS_SENTRUX_REQUIRED=true but no artifacts/sentrux-signal.yaml (set HARNESS_RUN_DIR) and .pi/harness/evals/smoke/sentrux-stub.json missing",
+			);
+		}
+		ok("Sentrux stub present (run signal absent — prefer artifacts/sentrux-signal.yaml from /harness-run)");
 	}
-	ok("Sentrux stub present");
 
 	const { code, out } = await runNodeScript(
 		join(ROOT, ".pi", "scripts", "sentrux-rules-sync.mjs"),
@@ -262,6 +301,11 @@ async function main() {
 	if (!policyGateSrc.includes('pi.on("tool_call", async (event, ctx)')) {
 		fail("policy-gate tool_call must receive ctx for run context");
 	}
+	if (!policyGateSrc.includes("evaluateContextModeMutation")) {
+		fail(
+			"policy-gate.ts must evaluate context-mode execute tools via evaluateContextModeMutation",
+		);
+	}
 	ok("policy-gate plan-phase writes");
 
 	const runCtxFixture = join(SMOKE, "run-context.fixture.json");
@@ -288,6 +332,7 @@ async function main() {
 	ok("test-diff-golden.json");
 
 	await checkSentruxGate();
+	await checkModelRouterThinkingOnly();
 
 	if (!(await fileExists(AGENTS_MANIFEST))) {
 		fail(

package/.pi/scripts/harness-web-policy-guard.mjs

@@ -0,0 +1,68 @@
+#!/usr/bin/env node
+/** Package-wide web-policy guard.
+ * Rejects raw HTTP shell/client paths unless they are in approved harness/API
+ * abstraction files. This is a static smoke guard; runtime blocking remains in
+ * .pi/extensions/harness-web-guard.ts.
+ */
+
+import { existsSync, readFileSync, readdirSync, statSync } from "node:fs";
+import { join, relative, resolve } from "node:path";
+
+const ROOT = resolve(new URL("../..", import.meta.url).pathname);
+const ALLOWED_FILES = new Set([
+	".pi/extensions/harness-web-guard.ts",
+	".pi/extensions/harness-web-tools.ts",
+	".pi/extensions/lib/harness-web/run-cli.ts",
+	".pi/extensions/harness-run-context.ts",
+	".pi/extensions/lib/ask-user/schema.ts",
+	".pi/scripts/harness-web.py",
+	".pi/scripts/harness-web-search.md",
+	".pi/scripts/harness-web-policy-guard.mjs",
+	".agents/skills/scrapling-web/SKILL.md",
+	".pi/scripts/harness-cli-verify.sh",
+	".pi/scripts/harness_web/output.py",
+	"AGENTS.md",
+]);
+const SKIP_DIRS = new Set([".git", "node_modules", "vendor", "graphify-out", "graphify-books-out", ".web", ".cocoindex_code", ".agents", ".cursor", "raw"]);
+const TEXT_EXTS = new Set([".js", ".mjs", ".ts", ".tsx", ".json", ".yaml", ".yml", ".py", ".sh", ".toml", ".example", ".template"]);
+const NEEDLES = [
+	{ name: "raw curl/wget URL", re: /\b(?:curl|wget)\b[^\n]*https?:\/\//i },
+	{ name: "raw Node HTTP client", re: /\b(?:fetch|request|get)\s*\(\s*[`'"]https?:\/\//i },
+	{ name: "raw Python HTTP client", re: /\brequests\.(?:get|post|put|delete|head)\s*\(\s*[`'"]https?:\/\//i },
+	{ name: "Firecrawl path", re: /\bfirecrawl\b/i },
+];
+
+function ext(path) {
+	const i = path.lastIndexOf(".");
+	return i >= 0 ? path.slice(i) : "";
+}
+
+function walk(dir, out = []) {
+	for (const name of readdirSync(dir)) {
+		const abs = join(dir, name);
+		const r = relative(ROOT, abs);
+		if (SKIP_DIRS.has(name) || r.startsWith(".pi/harness/runs/") || r.startsWith(".pi/agents/") || r.startsWith(".pi/skills/") || r.startsWith(".pi/prompts/") || r.startsWith(".pi/harness/docs/")) continue;
+		const st = statSync(abs);
+		if (st.isDirectory()) walk(abs, out);
+		else if (TEXT_EXTS.has(ext(name)) || name.includes("template") || name.includes("example")) out.push(abs);
+	}
+	return out;
+}
+
+const hits = [];
+for (const abs of walk(ROOT)) {
+	const r = relative(ROOT, abs);
+	if (ALLOWED_FILES.has(r)) continue;
+	let text = "";
+	try { text = readFileSync(abs, "utf8"); } catch { continue; }
+	for (const needle of NEEDLES) {
+		const m = text.match(needle.re);
+		if (m) hits.push({ file: r, rule: needle.name, match: m[0].slice(0, 140) });
+	}
+}
+
+if (hits.length) {
+	console.error(JSON.stringify({ ok: false, hits }, null, 2));
+	process.exit(1);
+}
+console.log(JSON.stringify({ ok: true, scanned_root: ROOT }, null, 2));

package/.pi/scripts/validate-plan-dag.mjs

@@ -12,9 +12,9 @@ import { readYamlFile, writeYamlFile } from "../lib/harness-yaml.mjs";
 const ROOT = join(dirname(fileURLToPath(import.meta.url)), "..", "..");
 
 const MINIMUMS = {
-	low: { phases: 2, work_items: 5, acceptance_checks: 3, risks: 0 },
-	med: { phases: 3, work_items: 8, acceptance_checks: 5, risks: 3 },
-	high: { phases: 4, work_items: 12, acceptance_checks: 8, risks: 3 },
+	low: { phases: 2, work_items: 2, acceptance_checks: 3, risks: 0 },
+	med: { phases: 3, work_items: 4, acceptance_checks: 5, risks: 3 },
+	high: { phases: 4, work_items: 6, acceptance_checks: 8, risks: 3 },
 };
 
 function fail(msg) {

package/AGENTS.md

@@ -10,6 +10,7 @@ Created: 2026-05-14
 - ./raw/ → Source documents for graphify ingestion
 - docs/adr/ → Repo-level Architectural Decision Records
 - .pi/harness/docs/adrs/ → Harness ADRs (team-shared; [index](.pi/harness/docs/adrs/README.md))
+- .pi/harness/docs/practice-map.md → Phase → practice → agent spawn topology for `/harness-plan`, `/harness-run`, `/harness-review`
 - .pi/skills/ → Agent skills
 - .pi/agents/ → Specialized agents
 

package/CHANGELOG.md

@@ -4,6 +4,28 @@ All notable changes to this project are documented in this file.
 
 ## [Unreleased]
 
+## [v0.18.0] — 2026-05-23
+
+### ✨ Features
+
+- **Harness:** Agent-native plan/review workflows (ADR 0042–0044) — path-first `approve_plan` / `create_plan` / `submit_*`, lakes on execution plans, `parallel_probes` debate, `review-outcome.yaml` routing, `/harness-steer` repair loop, `merge_harness_yaml`, and `harness_synthesize_repair_brief`.
+- **Harness:** Practice map, planning-context and plan-synthesizer agents, PostHog bootstrap, graphify KB updater corpus, and Sentrux steward prompt.
+
+### ✅ Tests
+
+- Add harness tool-payload, post-run routing, subprocess bootstrap, artifact gate, and topology tests; extend smoke plan fixture for parallel probes.
+
+## [v0.17.0] — 2026-05-22
+
+### ✨ Features
+
+- **Model router:** Session-locked model SKU at start (initial prompt + system prompt); per-turn routing adjusts thinking tier only; subagents lock from agent `systemPrompt` complexity.
+- **Harness:** Thinking-only profile shape in generator/verify; plan review gate, debate eligibility, and smoke fixture updates.
+
+### ✅ Tests
+
+- Add `harness-model-router-routing` and plan-debate eligibility coverage.
+
 ## [v0.16.0] — 2026-05-19
 
 ### ✨ Features

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "ultimate-pi",
-	"version": "0.16.0",
+	"version": "0.18.0",
 	"description": "Ultimate AI coding harness for pi.dev — extensible skills, Obsidian wiki knowledge layer, compressed context, deterministic output",
 	"keywords": [
 		"pi-package",
@@ -74,7 +74,7 @@
 		"@earendil-works/pi-coding-agent": "*"
 	},
 	"scripts": {
-		"check:ts": "tsc --noEmit --target ES2023 --lib ES2023 --moduleResolution nodenext --module nodenext --skipLibCheck .pi/extensions/custom-system-prompt.ts .pi/lib/harness-run-context.ts .pi/lib/harness-ui-state.ts .pi/extensions/harness-run-context.ts .pi/extensions/lib/harness-vcc-settings.ts .pi/extensions/dotenv-loader.ts .pi/extensions/lib/posthog-node.d.ts .pi/extensions/lib/harness-posthog.ts .pi/extensions/lib/harness-paths.ts .pi/extensions/pi-model-router-harness.ts .pi/extensions/provider-payload-sanitize.ts .pi/extensions/harness-telemetry.ts .pi/extensions/harness-ask-user.ts .pi/extensions/harness-plan-approval.ts .pi/extensions/lib/ask-user/schema.ts .pi/extensions/lib/ask-user/types.ts .pi/extensions/lib/ask-user/validate.ts .pi/extensions/lib/ask-user/dialog.ts .pi/extensions/lib/ask-user/fallback.ts .pi/extensions/lib/ask-user/render.ts .pi/extensions/lib/plan-approval/types.ts .pi/extensions/lib/plan-approval/schema.ts .pi/extensions/lib/plan-approval/validate.ts .pi/extensions/lib/plan-approval/format-plan.ts .pi/extensions/lib/plan-approval/dialog.ts .pi/extensions/lib/plan-approval/render.ts .pi/extensions/lib/plan-approval/create-plan.ts .pi/extensions/harness-subagents.ts .pi/extensions/lib/harness-subagents-bridge.ts .pi/extensions/lib/harness-cocoindex-refresh.ts .pi/extensions/lib/harness-subagent-auth.ts .pi/extensions/lib/harness-subagent-policy.ts .pi/extensions/lib/harness-subagent-precheck.ts .pi/extensions/lib/harness-spawn-budget.ts .pi/extensions/lib/spawn-policy.ts vendor/pi-subagents/src/agents.ts vendor/pi-subagents/src/subagents.ts .pi/extensions/trace-recorder.ts .pi/extensions/observation-bus.ts .pi/extensions/drift-monitor.ts .pi/extensions/policy-gate.ts .pi/extensions/budget-guard.ts .pi/extensions/debate-orchestrator.ts .pi/extensions/harness-debate-tools.ts .pi/extensions/lib/debate-bus-core.ts .pi/extensions/lib/debate-bus-state.ts .pi/extensions/lib/plan-debate-gate.ts .pi/extensions/lib/plan-debate-id.ts .pi/extensions/lib/plan-messenger.ts .pi/extensions/lib/plan-debate-envelope.ts .pi/extensions/lib/plan-review-integrator-rules.ts .pi/extensions/lib/plan-scope-guard.ts .pi/extensions/lib/plan-debate-write-guard.ts .pi/extensions/lib/plan-debate-lane.ts .pi/extensions/lib/plan-debate-round-status.ts .pi/extensions/harness-live-widget.ts .pi/extensions/sentrux-rules-sync.ts .pi/extensions/custom-header.ts .pi/extensions/harness-web-tools.ts .pi/extensions/harness-web-guard.ts .pi/extensions/lib/harness-web/run-cli.ts",
+		"check:ts": "tsc --noEmit --target ES2023 --lib ES2023 --moduleResolution nodenext --module nodenext --skipLibCheck .pi/extensions/custom-system-prompt.ts .pi/lib/harness-run-context.ts .pi/lib/harness-context-mode-policy.ts .pi/lib/harness-ui-state.ts .pi/extensions/harness-run-context.ts .pi/extensions/lib/harness-vcc-settings.ts .pi/extensions/dotenv-loader.ts .pi/extensions/00-posthog-network-bootstrap.ts .pi/extensions/lib/posthog-client.ts .pi/extensions/lib/posthog-node.d.ts .pi/extensions/lib/harness-posthog.ts .pi/extensions/lib/harness-paths.ts .pi/extensions/pi-model-router-harness.ts .pi/extensions/provider-payload-sanitize.ts .pi/extensions/harness-telemetry.ts .pi/extensions/harness-ask-user.ts .pi/extensions/harness-plan-approval.ts .pi/extensions/lib/ask-user/schema.ts .pi/extensions/lib/ask-user/types.ts .pi/extensions/lib/ask-user/validate.ts .pi/extensions/lib/ask-user/dialog.ts .pi/extensions/lib/ask-user/fallback.ts .pi/extensions/lib/ask-user/render.ts .pi/extensions/lib/plan-approval/types.ts .pi/extensions/lib/plan-approval/schema.ts .pi/extensions/lib/plan-approval/validate.ts .pi/extensions/lib/plan-approval/format-plan.ts .pi/extensions/lib/plan-approval/dialog.ts .pi/extensions/lib/plan-approval/render.ts .pi/extensions/lib/plan-approval/create-plan.ts .pi/extensions/harness-subagents.ts .pi/extensions/lib/harness-subagents-bridge.ts .pi/extensions/lib/harness-cocoindex-refresh.ts .pi/extensions/lib/harness-subagent-auth.ts .pi/extensions/lib/harness-subagent-policy.ts .pi/extensions/lib/harness-subagent-precheck.ts .pi/extensions/lib/harness-spawn-budget.ts .pi/extensions/lib/spawn-policy.ts vendor/pi-subagents/src/agents.ts vendor/pi-subagents/src/subagents.ts .pi/extensions/trace-recorder.ts .pi/extensions/observation-bus.ts .pi/extensions/drift-monitor.ts .pi/extensions/policy-gate.ts .pi/extensions/budget-guard.ts .pi/extensions/debate-orchestrator.ts .pi/extensions/harness-debate-tools.ts .pi/extensions/lib/debate-bus-core.ts .pi/extensions/lib/debate-bus-state.ts .pi/extensions/lib/plan-debate-gate.ts .pi/extensions/lib/plan-debate-id.ts .pi/extensions/lib/plan-messenger.ts .pi/extensions/lib/plan-debate-envelope.ts .pi/extensions/lib/plan-review-integrator-rules.ts .pi/extensions/lib/plan-scope-guard.ts .pi/extensions/lib/plan-debate-write-guard.ts .pi/extensions/lib/plan-debate-lane.ts .pi/extensions/lib/plan-debate-round-status.ts .pi/extensions/harness-live-widget.ts .pi/extensions/sentrux-rules-sync.ts .pi/extensions/custom-header.ts .pi/extensions/harness-web-tools.ts .pi/extensions/harness-web-guard.ts .pi/extensions/lib/harness-web/run-cli.ts",
 		"vendor:sync-router": "bash .pi/scripts/vendor-sync-pi-model-router.sh",
 		"vendor:sync-vcc": "bash .pi/scripts/vendor-sync-pi-vcc.sh",
 		"vendor:sync-subagents": "bash .pi/scripts/vendor-sync-pi-subagents.sh",
@@ -84,7 +84,7 @@
 		"format": "biome format --write",
 		"format:check": "biome format",
 		"prepare": "lefthook install",
-		"test": "node --test test/harness-verify.test.mjs test/harness-ask-user.test.mjs test/harness-subagents-loader.test.mjs test/harness-subagent-precheck.test.mjs test/sentrux-rules-sync.test.mjs test/harness-budget-guard.test.mjs && node .pi/harness/evals/smoke/smoke-harness-plan.mjs --fixture && npx -y tsx --test test/harness-vcc-settings.test.ts test/harness-live-widget-status.test.ts test/harness-plan-phase-policy.test.mjs test/harness-subagent-policy.test.mjs test/harness-spawn-budget.test.mjs test/harness-spawn-parse.test.mjs test/harness-schema-validate.test.mjs test/harness-turn-routing.test.mjs test/harness-budget-enforce.test.mjs test/harness-submit-policy.test.mjs test/plan-approval-format.test.mjs test/plan-approval-dialog.test.mjs test/plan-approval-sync.test.mjs test/plan-create-plan.test.mjs test/plan-review-format.test.mjs test/debate-plan-phase.test.mjs test/plan-debate-eligibility.test.mjs test/plan-messenger-gate.test.mjs test/plan-debate-lane-apply.test.mjs",
+		"test": "node --test test/harness-verify.test.mjs test/posthog-client.test.mjs test/harness-ask-user.test.mjs test/harness-subagents-loader.test.mjs test/harness-subagent-precheck.test.mjs test/sentrux-rules-sync.test.mjs test/harness-budget-guard.test.mjs && node .pi/harness/evals/smoke/smoke-harness-plan.mjs --fixture && npx -y tsx --test test/harness-vcc-settings.test.ts test/harness-run-context-postrun.test.mjs test/harness-tool-payload.test.mjs test/harness-live-widget-status.test.ts test/harness-plan-phase-policy.test.mjs test/harness-context-mode-policy.test.mjs test/harness-subprocess-bootstrap.test.mjs test/harness-subagent-policy.test.mjs test/harness-subagent-precheck-topology.test.mjs test/plan-approval-readiness.test.mjs test/harness-spawn-budget.test.mjs test/harness-spawn-parse.test.mjs test/harness-schema-validate.test.mjs test/harness-turn-routing.test.mjs test/harness-budget-enforce.test.mjs test/harness-submit-policy.test.mjs test/plan-approval-format.test.mjs test/plan-approval-dialog.test.mjs test/plan-approval-sync.test.mjs test/plan-create-plan.test.mjs test/plan-review-format.test.mjs test/debate-plan-phase.test.mjs test/plan-debate-eligibility.test.mjs test/plan-messenger-gate.test.mjs test/plan-debate-lane-apply.test.mjs",
 		"test:vcc": "npx -y tsx --test vendor/pi-vcc/tests/*.test.ts",
 		"harness:sentrux-bootstrap": "node .pi/scripts/harness-sentrux-bootstrap.mjs",
 		"harness:sentrux-sync": "node .pi/scripts/sentrux-rules-sync.mjs --force",
@@ -108,7 +108,8 @@
 		"croner": "^9.0.0",
 		"jimp": "^1.6.1",
 		"nanoid": "^5.1.5",
-		"posthog-node": "^5.30.6"
+		"posthog-node": "^5.30.6",
+		"undici": "^7.16.0"
 	},
 	"overrides": {
 		"@mariozechner/pi-agent-core": "npm:@earendil-works/pi-agent-core@0.74.1",

package/vendor/pi-model-router/UPSTREAM_PIN.md

@@ -3,6 +3,8 @@
 - **Repository:** https://github.com/yeliu84/pi-model-router
 - **License:** MIT (`LICENSE` in this tree)
 - **Pinned upstream commit:** `8c60095da0e753c242c4be9bb617b85f4dd3255c`
-- **Local changes:** TypeScript imports in `extensions/*.ts` use `@earendil-works/*`; relative imports end in `.js` for NodeNext; `package.json` peerDependencies list `@earendil-works/*`.
+- **Local changes:**
+  - TypeScript imports in `extensions/*.ts` use `@earendil-works/*`; relative imports end in `.js` for NodeNext; `package.json` peerDependencies list `@earendil-works/*`.
+  - **Session-locked routing (ultimate-pi harness):** one concrete model per session/profile, chosen from initial prompt + system prompt complexity; per-turn routing adjusts **thinking tier only** (`sessionLock` persisted in `router-state`). Harness generator emits the same `model` on high/medium/low tiers.
 
 **Refresh upstream:** run `npm run vendor:sync-router` from ultimate-pi root (updates this file with the latest commit SHA).

package/vendor/pi-model-router/extensions/commands.ts

@@ -58,12 +58,12 @@ export const registerCommands = (
   const SUBCOMMAND_DETAILS = [
     { name: 'status', desc: 'Show current router status' },
     { name: 'profile', desc: 'Switch to a different router profile' },
-    { name: 'pin', desc: 'Pin routing for a profile to a specific tier' },
+    { name: 'pin', desc: 'Pin thinking tier for a profile (model stays session-locked)' },
     { name: 'thinking', desc: 'Override thinking level for a tier or profile' },
     { name: 'disable', desc: 'Disable the router and restore last model' },
     {
       name: 'fix',
-      desc: 'Correct the last routing decision and pin that tier',
+      desc: 'Correct last thinking tier and pin it (model stays session-locked)',
     },
     { name: 'widget', desc: 'Toggle the router status widget' },
     { name: 'debug', desc: 'Toggle or clear router debug history' },
@@ -676,10 +676,10 @@ export const registerCommands = (
               'Router Subcommands:',
               '  status                      Show current status, profile, pin, cost, and last decision.',
               '  profile [name]              Switch to a profile (enables router if off). Lists available if no name.',
-              '  pin [profile] <tier|auto>   Force a tier (high|medium|low) for a profile or set to auto.',
+              '  pin [profile] <tier|auto>   Pin thinking tier (high|medium|low); model stays locked for the session.',
               '  thinking [prof] [tier] <lv> Override thinking level for a profile/tier (off|minimal|...|xhigh|auto).',
               '  disable                     Disable the router and restore the last used non-router model.',
-              '  fix <tier>                  Correct the last routing decision and pin that tier for the current profile.',
+              '  fix <tier>                  Correct last thinking tier and pin it for the current profile.',
               '  widget <on|off|toggle>      Control the persistent status widget visibility.',
               '  debug <on|off|show|clear>   Control routing debug logging to notifications and history.',
               '  reload                      Hot-reload the configuration JSON from .pi/model-router.json.',

package/vendor/pi-model-router/extensions/index.ts

@@ -10,6 +10,7 @@ import {
   type RouterThinkingByProfile,
   type RouterTier,
   type CustomSessionEntry,
+  type SessionLock,
 } from './types.js';
 import {
   FALLBACK_CONFIG,
@@ -47,6 +48,7 @@ const routerExtension = (pi: ExtensionAPI) => {
   let lastPersistedSnapshot: string | undefined;
   let isInitialized = false;
   let isInternalModelSwitch = false;
+  let sessionLock: SessionLock | undefined;
 
   const setModelInternally = async (
     model: NonNullable<ExtensionContext['model']>,
@@ -94,6 +96,7 @@ const routerExtension = (pi: ExtensionAPI) => {
       lastDecision,
       lastNonRouterModel,
       accumulatedCost,
+      sessionLock,
     );
     const snapshot = JSON.stringify({
       ...state,
@@ -127,6 +130,7 @@ const routerExtension = (pi: ExtensionAPI) => {
         accumulatedCost,
         widgetEnabled,
         currentConfig,
+        sessionLock,
       ),
     reloadConfig: (
       ctx?: ExtensionContext,
@@ -204,6 +208,7 @@ const routerExtension = (pi: ExtensionAPI) => {
       }
       selectedProfile = resolvedProfile;
       routerEnabled = true;
+      sessionLock = undefined;
       persistState();
       actions.updateStatus(ctx);
       return true;
@@ -253,6 +258,12 @@ const routerExtension = (pi: ExtensionAPI) => {
           set accumulatedCost(v) {
             accumulatedCost = v;
           },
+          get sessionLock() {
+            return sessionLock;
+          },
+          set sessionLock(v) {
+            sessionLock = v;
+          },
         },
         {
           persistState,
@@ -290,6 +301,7 @@ const routerExtension = (pi: ExtensionAPI) => {
         ? `${ctx.model.provider}/${ctx.model.id}`
         : lastNonRouterModel;
     lastDecision = undefined;
+    sessionLock = undefined;
 
     const entries = ctx.sessionManager.getBranch() as CustomSessionEntry[];
     const savedState = entries
@@ -322,6 +334,12 @@ const routerExtension = (pi: ExtensionAPI) => {
         : [];
       lastNonRouterModel = savedState.lastNonRouterModel ?? lastNonRouterModel;
       accumulatedCost = savedState.accumulatedCost ?? 0;
+      if (
+        savedState.sessionLock &&
+        savedState.sessionLock.profile === selectedProfile
+      ) {
+        sessionLock = { ...savedState.sessionLock };
+      }
     }
 
     await actions.ensureValidActiveRouterProfile(ctx);
@@ -432,6 +450,9 @@ const routerExtension = (pi: ExtensionAPI) => {
       }
 
       routerEnabled = true;
+      if (selectedProfile !== profileName) {
+        sessionLock = undefined;
+      }
       selectedProfile = profileName;
     } else {
       routerEnabled = false;