@opengsd/gsd-pi 1.1.1-dev.b2556262 → 1.2.0-dev.4813ead6

package/src/resources/extensions/gsd/roadmap-slices.ts

@@ -189,9 +189,15 @@ export function parseRoadmapSlices(content: string): RoadmapSliceEntry[] {
       const risk = (riskMatch ? riskMatch[1] : "low") as RiskLevel;
 
       const depsMatch = rest.match(/`depends:\[([^\]]*)\]`/);
-      const depends = depsMatch && depsMatch[1]!.trim()
-        ? expandDependencies(depsMatch[1]!.split(",").map(s => s.trim()))
+      // Recovery fallback: double-bracket form `[[id]]` from serialized bracket-wrapped IDs
+      const fallbackDepsMatch = depsMatch ? null : rest.match(/`depends:\[(\[(?:[^\]]*)\](?:,\[(?:[^\]]*)\])*)\]`/);
+      const rawDepContent = (depsMatch ?? fallbackDepsMatch)?.[1] ?? "";
+      const SLICE_ID_RE = /^[A-Za-z0-9][A-Za-z0-9-]*$/;
+      const RANGE_RE = /^[A-Za-z]+\d+(?:-|\.\.)[A-Za-z]+\d+$/;
+      const rawDepParts = rawDepContent.trim()
+        ? rawDepContent.replace(/\[|\]/g, "").split(",").map(s => s.trim()).filter(s => SLICE_ID_RE.test(s) || RANGE_RE.test(s))
         : [];
+      const depends = expandDependencies(rawDepParts);
 
       // ADR-011: the renderer writes a `[sketch]` badge for sketch slices.
       // Parse it back so the is_sketch flag survives a markdown → DB re-import

package/src/resources/extensions/gsd/schemas/parsers.ts

@@ -67,7 +67,12 @@ export interface ParsedRoadmap {
 const TEMPLATE_TOKEN_RE = /\{\{[^}]+\}\}/;
 const H2_RE = /^##\s+(.+)$/gm;
 const H3_RE = /^###\s+(.+)$/gm;
-const MILESTONE_LINE_RE = /^-\s+\[([ x])\]\s+(M\d{3}):\s+(.+?)\s+(?:—|--|-)\s+(.+)$/gm;
+// A milestone line is single-line by construction. Every inter-token gap uses
+// horizontal-whitespace classes (`[^\S\n]`) rather than `\s`, because `\s`
+// matches newlines: a line missing a valid separator would otherwise let the
+// `\s+(?:—|--|-)\s+` clause "bridge" onto the NEXT bullet's `- `, consuming it
+// as the separator and silently swallowing the following well-formed milestone.
+const MILESTONE_LINE_RE = /^-[^\S\n]+\[([ x])\][^\S\n]+(M\d{3}):[^\S\n]+(.+?)[^\S\n]+(?:—|--|-)[^\S\n]+(.+)$/gm;
 const SLICE_HEADER_RE = /^###\s+(S\d{2})\s*(?:—|--|-)\s+(.+)$/m;
 const REQUIREMENT_HEADER_RE = /^###\s+(R\d{3})\s*(?:—|--|-)\s+(.+)$/m;
 

package/src/resources/extensions/gsd/slice-parallel-orchestrator.ts

@@ -30,7 +30,8 @@ import { isAbsolute, join, dirname, resolve } from "node:path";
 import { fileURLToPath } from "node:url";
 import { gsdRoot } from "./paths.js";
 import { createWorktree, worktreePath, removeWorktree } from "./worktree-manager.js";
-import { autoWorktreeBranch, runWorktreePostCreateHook, syncGsdStateToWorktree } from "./auto-worktree.js";
+import { autoWorktreeBranch, runWorktreePostCreateHook, syncGsdStateToWorktreeByScope } from "./auto-worktree.js";
+import { createWorkspace, scopeMilestone } from "./workspace.js";
 import {
   writeSessionStatus,
   removeSessionStatus,
@@ -163,7 +164,10 @@ function createSliceWorktree(basePath: string, milestoneId: string, sliceId: str
   if (hookError) {
     throw new Error(`slice worktree post-create hook failed (${wtName}): ${hookError}`);
   }
-  syncGsdStateToWorktree(basePath, wtPath);
+  syncGsdStateToWorktreeByScope(
+    scopeMilestone(createWorkspace(basePath), milestoneId),
+    scopeMilestone(createWorkspace(wtPath), milestoneId),
+  );
 
   if (!existsSync(join(wtPath, ".gsd"))) {
     throw new Error(`slice worktree preflight failed (${wtName}): missing .gsd in worktree`);

package/src/resources/extensions/gsd/state-reconciliation/drift/artifact-db.ts

@@ -335,21 +335,42 @@ function detectDiskSliceIdDivergenceForMilestone(
   return drifts;
 }
 
+type ArtifactDbDrift =
+  | DiskSliceIdDivergenceDrift
+  | ArtifactDbStatusDivergenceDrift
+  | CompletedMilestoneReopenedDrift;
+
+// #442 Phase 1.6: the three artifact/DB drift handlers (disk-slice-id,
+// artifact-db-status, completed-milestone-reopened) each call
+// detectArtifactDbDrift and then filter for their own kind — so the full
+// milestone→slice→task walk + artifact SQL + disk scan would run THREE times
+// per detection pass and discard 2/3 of the work. Memoize the result per
+// DriftContext so the three handlers share one computation. The key is the
+// ctx object, which detectAllDrift rebuilds for every pass (and which is
+// unreferenced once the pass ends, so the WeakMap entry is GC'd) — DB/disk
+// state is immutable within a single pass (repairs run only after detection),
+// so this is behavior-preserving. A fresh ctx (e.g. the maintenance command's
+// inline { basePath, state }) always recomputes.
+const _artifactDbDriftMemo = new WeakMap<DriftContext, ArtifactDbDrift[]>();
+
 export function detectArtifactDbDrift(
+  state: GSDState,
+  ctx: DriftContext,
+): ArtifactDbDrift[] {
+  const cached = _artifactDbDriftMemo.get(ctx);
+  if (cached) return cached;
+  const computed = computeArtifactDbDrift(state, ctx);
+  _artifactDbDriftMemo.set(ctx, computed);
+  return computed;
+}
+
+function computeArtifactDbDrift(
   _state: GSDState,
   ctx: DriftContext,
-): Array<
-  | DiskSliceIdDivergenceDrift
-  | ArtifactDbStatusDivergenceDrift
-  | CompletedMilestoneReopenedDrift
-> {
+): ArtifactDbDrift[] {
   if (!isDbAvailable()) return [];
 
-  const drifts: Array<
-    | DiskSliceIdDivergenceDrift
-    | ArtifactDbStatusDivergenceDrift
-    | CompletedMilestoneReopenedDrift
-  > = [];
+  const drifts: ArtifactDbDrift[] = [];
 
   for (const milestone of getAllMilestones()) {
     if (isClosedStatus(milestone.status)) continue;

package/src/resources/extensions/gsd/state.ts

@@ -52,7 +52,6 @@ import { isTerminalMilestoneSummaryContent } from './milestone-summary-classifie
 
 import {
   isDbAvailable,
-  wasDbOpenAttempted,
   getAllMilestones,
   getMilestone,
   getMilestoneSlices,
@@ -63,9 +62,14 @@ import {
   getLatestAssessmentByScope,
   getPendingGateCountForTurn,
 } from './gsd-db.js';
+import { wasWorkflowDatabaseOpenAttempted } from './db-workspace.js';
 import { formatCompletePhaseNextAction, countUnmappedActiveRequirements } from './requirements-backlog.js';
 import type { MilestoneRow } from './db-milestone-artifact-rows.js';
 import type { SliceRow, TaskRow } from './db-task-slice-rows.js';
+import {
+  classifyMilestoneReadiness,
+  readinessNeedsDiscussion,
+} from './milestone-readiness.js';
 
 function formatNeedsAttentionBlocker(milestoneId: string): string {
   return [
@@ -395,7 +399,7 @@ export async function deriveState(
     stopDbTimer({ phase: result.phase, milestone: result.activeMilestone?.id });
     _telemetry.dbDeriveCount++;
   } else {
-    if (wasDbOpenAttempted()) {
+    if (wasWorkflowDatabaseOpenAttempted()) {
       logWarning("state", "DB unavailable — refusing implicit markdown state derivation");
     }
     result = {
@@ -507,6 +511,12 @@ async function buildRegistryAndFindActive(
     const title = stripMilestonePrefix(m.title) || m.id;
     const hasContext = !!resolveMilestoneFile(basePath, m.id, "CONTEXT");
     const hasDraftContext = !hasContext && !!resolveMilestoneFile(basePath, m.id, "CONTEXT-DRAFT");
+    const readiness = classifyMilestoneReadiness({
+      status: m.status,
+      hasContext,
+      hasDraftContext,
+      sliceCount: slices.length,
+    });
 
     if (!activeMilestoneFound) {
       const deps = m.depends_on;
@@ -517,9 +527,9 @@ async function buildRegistryAndFindActive(
         continue;
       }
 
-      if (m.status === 'queued' && slices.length === 0 && !hasContext) {
+      if (readiness.kind === 'queued-shell') {
         if (!firstDeferredQueuedShell) {
-          firstDeferredQueuedShell = { id: m.id, title, deps, hasDraftContext };
+          firstDeferredQueuedShell = { id: m.id, title, deps, hasDraftContext: readiness.hasDraftContext };
         }
         registry.push({ id: m.id, title, status: 'pending', ...(deps.length > 0 ? { dependsOn: deps } : {}) });
         continue;
@@ -533,7 +543,7 @@ async function buildRegistryAndFindActive(
         continue;
       }
 
-      if ((m.status === 'needs-discussion' && !hasContext) || hasDraftContext) activeMilestoneHasDraft = true;
+      if (readinessNeedsDiscussion(readiness)) activeMilestoneHasDraft = true;
 
       activeMilestone = { id: m.id, title };
       activeMilestoneSlices = slices;

package/src/resources/extensions/gsd/templates/plan.md

@@ -46,6 +46,13 @@
      For simple projects or scripts:
      - Executable verification commands (bash assertions, curl checks, etc.) are sufficient
 
+     For browser-facing web apps (React, Next, Vue, Vite, static HTML, etc.):
+     - Prefer `npx playwright test <spec>` in Verification when `@playwright/test` exists
+     - If Playwright is not set up yet, the first UI slice should add `playwright.config.ts`,
+       a minimal smoke spec (for example `e2e/smoke.spec.ts`), and a safe verify command
+     - Slice closeout UAT must declare `browser-executable` or `runtime-executable` — not
+       `artifact-driven` — when checks navigate to localhost or exercise the live UI
+
      If the project has no test framework and the work is non-trivial,
      the first task should set one up. A test runner costs 2 minutes
      and pays for itself immediately.

package/src/resources/extensions/gsd/templates/project.md

@@ -20,6 +20,7 @@
 
 - **Complexity:** {{simple | complex}}
 - **Why:** {{one-line rationale citing the signals that decided it}}
+- **Web stack:** {{browser-facing framework or "not a web UI" — e.g. Next.js + Playwright, static HTML, N/A for CLI/API-only}}
 
 ## Current State
 

package/src/resources/extensions/gsd/templates/roadmap.md

@@ -42,7 +42,7 @@
 - Contract verification: {{tests / shell verifiers / fixtures / artifact checks}}
 - Integration verification: {{real subsystem interaction that must be exercised, or none}}
 - Operational verification: {{service lifecycle / restart / reconnect / supervision / deploy-install behavior, or none}}
-- UAT / human verification: {{what needs real human judgment, or none}}
+- UAT / human verification: {{what needs real human judgment, or none — for web apps prefer browser-executable / runtime-executable Playwright checks over deferring UI proof to humans}}
 
 ## Milestone Definition of Done
 

package/src/resources/extensions/gsd/templates/uat.md

@@ -5,9 +5,13 @@
 
 ## UAT Type
 
-- UAT mode: {{artifact-driven | live-runtime | human-experience | mixed}}
+- UAT mode: {{artifact-driven | browser-executable | runtime-executable | live-runtime | human-experience | mixed}}
 - Why this mode is sufficient: {{reason}}
 
+<!-- Web apps: never use artifact-driven when steps open a browser, navigate to localhost,
+     or capture screenshots. Use browser-executable (browser_* tools) or runtime-executable
+     (npx playwright test). Name dev-server preconditions below. -->
+
 ## Preconditions
 
 {{whatMustBeTrueBeforeTesting — server running, data seeded, etc.}}

package/src/resources/extensions/gsd/tests/artifact-db-drift-memo.test.ts

@@ -0,0 +1,66 @@
+// Project/App: gsd-pi
+// File Purpose: #442 Phase 1.6 — detectArtifactDbDrift is memoized per
+// DriftContext so the three artifact/DB drift handlers share one
+// milestone→slice→task walk per detection pass instead of recomputing it
+// three times. Asserts external behavior: same ctx returns the same result
+// (shared within a pass); a fresh ctx recomputes identical content (no
+// cross-pass leakage).
+
+import test from "node:test";
+import assert from "node:assert/strict";
+import { mkdtempSync, mkdirSync, rmSync, writeFileSync } from "node:fs";
+import { join } from "node:path";
+import { tmpdir } from "node:os";
+
+import { openDatabase, closeDatabase, insertMilestone, insertSlice } from "../gsd-db.ts";
+import { detectArtifactDbDrift } from "../state-reconciliation/drift/artifact-db.ts";
+import type { DriftContext } from "../state-reconciliation/types.ts";
+import type { GSDState } from "../types.ts";
+
+function stubState(): GSDState {
+  return {
+    activeMilestone: { id: "M001", title: "M" },
+    activeSlice: null,
+    activeTask: null,
+    phase: "executing",
+    recentDecisions: [],
+    blockers: [],
+    nextAction: "",
+    registry: [],
+    requirements: { active: 0, validated: 0, deferred: 0, outOfScope: 0, blocked: 0, total: 0 },
+    progress: { milestones: { done: 0, total: 1 } },
+  };
+}
+
+test("#442: detectArtifactDbDrift is memoized per DriftContext", (t) => {
+  const base = mkdtempSync(join(tmpdir(), "gsd-artifact-memo-"));
+  const sliceDir = join(base, ".gsd", "milestones", "M001", "slices", "S01");
+  mkdirSync(sliceDir, { recursive: true });
+  t.after(() => {
+    try { closeDatabase(); } catch { /* noop */ }
+    rmSync(base, { recursive: true, force: true });
+  });
+
+  openDatabase(join(base, ".gsd", "gsd.db"));
+  insertMilestone({ id: "M001", title: "M", status: "active" });
+  insertSlice({ id: "S01", milestoneId: "M001", title: "Slice", status: "pending", risk: "low", depends: [], sequence: 1 });
+  // A SUMMARY on disk while the slice is still pending = artifact/DB divergence.
+  writeFileSync(join(sliceDir, "S01-SUMMARY.md"), "# S01 Summary\n");
+
+  const state = stubState();
+
+  const ctx1: DriftContext = { basePath: base, state };
+  const first = detectArtifactDbDrift(state, ctx1);
+  const firstAgain = detectArtifactDbDrift(state, ctx1);
+
+  // Same ctx → cache hit → identical array reference (the 3 handlers in one
+  // pass share this exact result).
+  assert.strictEqual(firstAgain, first, "same ctx must return the memoized result");
+  assert.ok(first.length > 0, "fixture should produce at least one drift record");
+
+  // Fresh ctx → recomputed (distinct instance) but identical content.
+  const ctx2: DriftContext = { basePath: base, state };
+  const second = detectArtifactDbDrift(state, ctx2);
+  assert.notStrictEqual(second, first, "a fresh ctx must recompute, not reuse the prior pass");
+  assert.deepEqual(second, first, "recomputed result must be identical content");
+});

package/src/resources/extensions/gsd/tests/ask-user-questions-render.test.ts

@@ -0,0 +1,92 @@
+/**
+ * Regression tests for ask_user_questions renderResult when Claude Code MCP
+ * omits structuredContent and the chat controller passes details as undefined
+ * or an empty object.
+ */
+
+import assert from "node:assert/strict";
+import { test } from "node:test";
+import askUserQuestionsExtension from "../../ask-user-questions.ts";
+
+const fakeTheme = {
+	fg: (_color: string, text: string) => text,
+	bold: (text: string) => text,
+};
+
+function getAskUserQuestionsTool() {
+	const tools: any[] = [];
+	askUserQuestionsExtension({ registerTool: (tool: any) => tools.push(tool) } as any);
+	const tool = tools.find((t) => t.name === "ask_user_questions");
+	assert.ok(tool, "ask_user_questions should be registered");
+	return tool;
+}
+
+function renderText(tool: any, result: unknown, args?: unknown): string {
+	const rendered = tool.renderResult(result, {}, fakeTheme, { args });
+	return String((rendered as any).content ?? (rendered as any).text ?? rendered);
+}
+
+const depthQuestion = {
+	id: "depth_check",
+	header: "Depth Check",
+	question: "Did I capture the depth right?",
+	options: [
+		{ label: "Yes, you got it", description: "Proceed" },
+		{ label: "Not quite - let me clarify", description: "Clarify" },
+	],
+};
+
+test("ask_user_questions renderResult shows answers when details is undefined (MCP text-only wire)", () => {
+	const tool = getAskUserQuestionsTool();
+	const contentText = JSON.stringify({
+		answers: { depth_check: { answers: ["Yes, you got it"] } },
+	});
+	const text = renderText(
+		tool,
+		{
+			content: [{ type: "text", text: contentText }],
+			details: undefined,
+			isError: false,
+		},
+		{ questions: [depthQuestion] },
+	);
+
+	assert.match(text, /✓/);
+	assert.match(text, /Depth Check/);
+	assert.match(text, /Yes, you got it/);
+	assert.doesNotMatch(text, /Cancelled/i);
+});
+
+test("ask_user_questions renderResult shows answers when details is empty object (#cc-elicitation)", () => {
+	const tool = getAskUserQuestionsTool();
+	const contentText = JSON.stringify({
+		answers: { depth_check: { answers: ["Yes, you got it"] } },
+	});
+	const text = renderText(
+		tool,
+		{
+			content: [{ type: "text", text: contentText }],
+			details: {},
+			isError: false,
+		},
+		{ questions: [depthQuestion] },
+	);
+
+	assert.match(text, /Yes, you got it/);
+	assert.doesNotMatch(text, /Cancelled/i);
+});
+
+test("ask_user_questions renderResult still shows Cancelled for explicit cancel payload", () => {
+	const tool = getAskUserQuestionsTool();
+	const text = renderText(
+		tool,
+		{
+			content: [{ type: "text", text: "ask_user_questions was cancelled before receiving a response" }],
+			details: { questions: [depthQuestion], response: null, cancelled: true },
+			isError: false,
+		},
+		{ questions: [depthQuestion] },
+	);
+
+	assert.match(text, /Cancelled/i);
+});

package/src/resources/extensions/gsd/tests/auto-dashboard.test.ts

@@ -29,11 +29,16 @@ import {
   _refreshLastCommitForTests,
   _getLastCommitForTests,
   _getLastCommitFetchedAtForTests,
+  formatToolSurfaceSnapshot,
   formatRuntimeHealthSignal,
   shouldRenderRoadmapProgress,
 } from "../auto-dashboard.ts";
 import { getAutoDashboardData } from "../auto.ts";
-import { autoSession } from "../auto-runtime-state.ts";
+import {
+  autoSession,
+  clearAutoToolSurfaceSnapshot,
+  recordAutoToolSurfaceSnapshot,
+} from "../auto-runtime-state.ts";
 import { formatRtkSavingsLabel } from "../../shared/rtk-session-stats.ts";
 import {
   openDatabase,
@@ -530,6 +535,29 @@ test("getAutoDashboardData returns RTK savings in the dashboard payload", () =>
   }
 });
 
+test("getAutoDashboardData exposes typed tool-surface snapshots", () => {
+  autoSession.reset();
+  clearAutoToolSurfaceSnapshot();
+  autoSession.active = true;
+  recordAutoToolSurfaceSnapshot({
+    source: "provider-adjustment",
+    unitType: "run-uat",
+    modelFacingToolNames: ["read"],
+    registeredToolNames: ["read", "browser_navigate"],
+    scopedToolNames: ["read", "browser_navigate"],
+    presentedToolNames: ["browser_navigate"],
+    capturedAt: 456,
+  });
+
+  const data = getAutoDashboardData();
+
+  assert.equal(data.toolSurface?.source, "provider-adjustment");
+  assert.equal(formatToolSurfaceSnapshot(data.toolSurface), "run-uat: model 1 / registered 2 / scoped 2 / presented 1");
+
+  autoSession.reset();
+  clearAutoToolSurfaceSnapshot();
+});
+
 test("RTK savings label formats the dashboard footer text", () => {
   assert.equal(formatRtkSavingsLabel(null), null);
   assert.equal(

package/src/resources/extensions/gsd/tests/auto-dispatch-baseline-harness.test.ts

@@ -0,0 +1,53 @@
+// Project/App: gsd-pi
+// File Purpose: Cover the getDebugCounters() snapshot getter added for the
+// per-dispatch benchmark harness (issue #442, Phase 0.3). The harness itself
+// (scripts/auto-dispatch-baseline.mjs) reads counters via this getter without
+// disabling debug (which is what writeDebugSummary does).
+
+import { test } from 'node:test';
+import assert from 'node:assert';
+import { mkdtempSync, mkdirSync } from 'node:fs';
+import { join } from 'node:path';
+import { tmpdir } from 'node:os';
+
+import {
+  enableDebug,
+  disableDebug,
+  debugCount,
+  getDebugCounters,
+} from '../debug-logger.ts';
+
+function tmpGsd(): string {
+  const tmp = mkdtempSync(join(tmpdir(), 'gsd-baseline-harness-'));
+  mkdirSync(join(tmp, '.gsd'), { recursive: true });
+  return tmp;
+}
+
+test('getDebugCounters returns a live snapshot of the hot-path counters', () => {
+  enableDebug(tmpGsd());
+
+  debugCount('deriveStateCalls', 3);
+  debugCount('parseRoadmapCalls');
+  debugCount('parsePlanCalls', 2);
+  debugCount('gitInvocations', 7);
+
+  const snap = getDebugCounters();
+  assert.strictEqual(snap.deriveStateCalls, 3);
+  assert.strictEqual(snap.parseRoadmapCalls, 1);
+  assert.strictEqual(snap.parsePlanCalls, 2);
+  assert.strictEqual(snap.gitInvocations, 7);
+
+  disableDebug();
+});
+
+test('getDebugCounters returns a copy — callers cannot mutate internal state', () => {
+  enableDebug(tmpGsd());
+  debugCount('gitInvocations', 5);
+
+  const snap = getDebugCounters() as Record<string, number>;
+  snap.gitInvocations = 999;
+
+  assert.strictEqual(getDebugCounters().gitInvocations, 5, 'internal counter must be unaffected by mutating the snapshot');
+
+  disableDebug();
+});