gsd-pi 2.76.0-dev.fe143342a → 2.77.0-dev.538325aea

package/dist/resources/extensions/gsd/bootstrap/register-hooks.js

@@ -13,12 +13,12 @@ import { loadToolApiKeys } from "../commands-config.js";
 import { loadFile, saveFile, formatContinue } from "../files.js";
 import { deriveState } from "../state.js";
 import { getAutoDashboardData, isAutoActive, isAutoPaused, markToolEnd, markToolStart, recordToolInvocationError } from "../auto.js";
-import { hideFooter } from "../auto-dashboard.js";
 import { isParallelActive, shutdownParallel } from "../parallel-orchestrator.js";
 import { checkToolCallLoop, resetToolCallLoopGuard } from "./tool-call-loop-guard.js";
 import { saveActivityLog } from "../activity-log.js";
 import { resetAskUserQuestionsCache } from "../../ask-user-questions.js";
-import { recordToolCall as safetyRecordToolCall, recordToolResult as safetyRecordToolResult } from "../safety/evidence-collector.js";
+import { recordToolCall as safetyRecordToolCall, recordToolResult as safetyRecordToolResult, saveEvidenceToDisk } from "../safety/evidence-collector.js";
+import { parseUnitId } from "../unit-id.js";
 import { classifyCommand } from "../safety/destructive-guard.js";
 import { logWarning as safetyLogWarning } from "../workflow-logger.js";
 import { installNotifyInterceptor } from "./notify-interceptor.js";
@@ -37,7 +37,9 @@ export function registerHooks(pi, ecosystemHandlers) {
         initNotificationStore(process.cwd());
         installNotifyInterceptor(ctx);
         initNotificationWidget(ctx);
-        initHealthWidget(ctx);
+        if (!isAutoActive()) {
+            initHealthWidget(ctx);
+        }
         resetWriteGateState();
         resetToolCallLoopGuard();
         resetAskUserQuestionsCache();
@@ -79,7 +81,7 @@ export function registerHooks(pi, ecosystemHandlers) {
         }
         loadToolApiKeys();
         if (isAutoActive()) {
-            ctx.ui.setFooter(hideFooter);
+            ctx.ui.setWidget("gsd-health", undefined);
         }
     });
     pi.on("session_switch", async (_event, ctx) => {
@@ -101,7 +103,7 @@ export function registerHooks(pi, ecosystemHandlers) {
         }
         loadToolApiKeys();
         if (isAutoActive()) {
-            ctx.ui.setFooter(hideFooter);
+            ctx.ui.setWidget("gsd-health", undefined);
         }
     });
     pi.on("before_agent_start", async (event, ctx) => {
@@ -441,7 +443,7 @@ export function registerHooks(pi, ecosystemHandlers) {
     pi.on("tool_execution_start", async (event) => {
         if (!isAutoActive())
             return;
-        markToolStart(event.toolCallId);
+        markToolStart(event.toolCallId, event.toolName);
     });
     pi.on("tool_execution_end", async (event) => {
         markToolEnd(event.toolCallId);
@@ -456,6 +458,15 @@ export function registerHooks(pi, ecosystemHandlers) {
         // Safety harness: record tool execution results for evidence cross-referencing
         if (isAutoActive()) {
             safetyRecordToolResult(event.toolCallId, event.toolName, event.result, event.isError);
+            // Persist evidence to disk after each tool result so it survives a session
+            // restart mid-unit (Bug #4385 — non-persisted evidence false positives).
+            const dash = getAutoDashboardData();
+            if (dash.basePath && dash.currentUnit?.type === "execute-task") {
+                const { milestone: pMid, slice: pSid, task: pTid } = parseUnitId(dash.currentUnit.id);
+                if (pMid && pSid && pTid) {
+                    saveEvidenceToDisk(dash.basePath, pMid, pSid, pTid);
+                }
+            }
         }
     });
     pi.on("model_select", async (_event, ctx) => {

package/dist/resources/extensions/gsd/bootstrap/write-gate.js

@@ -24,8 +24,29 @@ const QUEUE_SAFE_TOOLS = new Set([
 /**
  * Bash commands that are read-only / investigative — safe during queue mode.
  * Matches the leading command in a bash invocation.
+ *
+ * Extension policy: add commands here when they are read-only / diagnostic.
+ * Never add commands that mutate project state (write files, run builds that
+ * emit artifacts, install packages, etc.).
+ *
+ * Current read-only additions (Bug #4385):
+ *   npm run <diagnostic> — read-only diagnostic scripts: test, lint, typecheck, etc.
+ *                         NOT: build, install, compile, generate, deploy (artifact-producing)
+ *   npm ls/list/info    — inspect installed packages (read-only)
+ *   npm outdated/audit  — security/update checks (read-only)
+ *   npx <pkg>           — run a package binary without installing globally
+ *   tsx                 — TypeScript runner used for dry-run / inspection scripts
+ *   node --print        — evaluate and print an expression, no side effects
+ *   python / python3    — script inspection, version checks
+ *   pip / pip3 show     — show installed package info (read-only)
+ *   jq                  — read-only JSON query
+ *   yq                  — read-only YAML query
+ *   curl -s / curl --silent — fetch for inspection (no -o / no output redirect)
+ *   openssl version     — version / certificate inspection
+ *   env / printenv      — print environment variables
+ *   true / false        — shell no-ops / test exit codes
  */
-const BASH_READ_ONLY_RE = /^\s*(cat|head|tail|less|more|wc|file|stat|du|df|which|type|echo|printf|ls|find|grep|rg|awk|sed\b(?!.*-i)|sort|uniq|diff|comm|tr|cut|tee\s+-a\s+\/dev\/null|git\s+(log|show|diff|status|branch|tag|remote|rev-parse|ls-files|blame|shortlog|describe|stash\s+list|config\s+--get|cat-file)|gh\s+(issue|pr|api|repo|release)\s+(view|list|diff|status|checks)|mkdir\s+-p\s+\.gsd|rtk\s)/;
+const BASH_READ_ONLY_RE = /^\s*(cat|head|tail|less|more|wc|file|stat|du|df|which|type|echo|printf|ls|find|grep|rg|awk|sed\b(?!.*-i)|sort|uniq|diff|comm|tr|cut|tee\s+-a\s+\/dev\/null|git\s+(log|show|diff|status|branch|tag|remote|rev-parse|ls-files|blame|shortlog|describe|stash\s+list|config\s+--get|cat-file)|gh\s+(issue|pr|api|repo|release)\s+(view|list|diff|status|checks)|mkdir\s+-p\s+\.gsd|rtk\s|npm\s+run\s+(test|test:\w+|lint|lint:\w+|typecheck|type-check|type-check:\w+|check|verify|audit|outdated|format:check|ci|validate)\b|npm\s+(ls|list|info|view|show|outdated|audit|explain|doctor|ping|--version|-v)\b|npx\s|tsx\s|node\s+(--print|--version|-v\b)|python[23]?\s+(-c\s+'[^']*'|--version|-V\b|-m\s+(pip\s+show|pip\s+list|site))|pip[23]?\s+(show|list|freeze|check|index\s+versions)\b|jq\s|yq\s|curl\s+(-s\b|--silent\b)(?!\s+[^|>]*\s-[oO]\b)(?!\s+[^|>]*\s--output\b)[^|>]*$|openssl\s+(version|x509|s_client)|env\b|printenv\b|true\b|false\b)/;
 const verifiedDepthMilestones = new Set();
 let activeQueuePhase = false;
 /**
@@ -99,10 +120,21 @@ function normalizeWriteGateSnapshot(value) {
         pendingGateId: typeof record.pendingGateId === "string" ? record.pendingGateId : null,
     };
 }
+const EMPTY_SNAPSHOT = {
+    verifiedDepthMilestones: [],
+    activeQueuePhase: false,
+    pendingGateId: null,
+};
 export function loadWriteGateSnapshot(basePath = process.cwd()) {
     const path = writeGateSnapshotPath(basePath);
-    if (!existsSync(path))
+    if (!existsSync(path)) {
+        // When persist mode is active and the file has been deleted, treat it as a
+        // full state reset so deleting the file clears the HARD BLOCK gate.
+        // In non-persist mode the file is never written, so fall back to in-memory.
+        if (shouldPersistWriteGateSnapshot())
+            return EMPTY_SNAPSHOT;
         return currentWriteGateSnapshot();
+    }
     try {
         return normalizeWriteGateSnapshot(JSON.parse(readFileSync(path, "utf-8")));
     }

package/dist/resources/extensions/gsd/clean-root-preflight.js

@@ -0,0 +1,93 @@
+/**
+ * clean-root-preflight.ts — Preflight gate for dirty working trees before milestone merges.
+ *
+ * #2909: Adds a fast-path git status check before milestone completion merges.
+ * When the working tree is dirty the user is warned and changes are auto-stashed
+ * so the merge can proceed cleanly.  After the merge completes, postflightPopStash
+ * restores the stashed changes.
+ *
+ * Design constraints (from Trek-e approval):
+ *  - Warn the user before stashing (no silent surprises)
+ *  - git stash push / git stash pop only — no custom stash management layer
+ *  - Stash/pop errors are logged but MUST NOT block the merge
+ *  - Fast-path status check — clean trees pay no extra cost
+ */
+import { execFileSync } from "node:child_process";
+import { GIT_NO_PROMPT_ENV } from "./git-constants.js";
+import { logWarning } from "./workflow-logger.js";
+import { nativeHasChanges } from "./native-git-bridge.js";
+/**
+ * Check the working tree for dirty files before a milestone merge.
+ *
+ * Clean tree path: O(1) — returns immediately with stashPushed=false.
+ *
+ * Dirty tree path:
+ *  1. Emits a warning notification via the provided `notify` callback.
+ *  2. Runs `git stash push --include-untracked -m "gsd-preflight-stash"`.
+ *  3. Returns stashPushed=true so the caller knows to call postflightPopStash.
+ *
+ * Any stash error is logged but does NOT throw — the merge proceeds regardless.
+ */
+export function preflightCleanRoot(basePath, milestoneId, notify) {
+    // Fast-path: clean tree — nothing to do
+    let isDirty = false;
+    try {
+        isDirty = nativeHasChanges(basePath);
+    }
+    catch (err) {
+        // If the status check itself fails, treat as clean and let the merge decide
+        logWarning("preflight", `clean-root status check failed: ${err instanceof Error ? err.message : String(err)}`);
+        return { stashPushed: false, summary: "" };
+    }
+    if (!isDirty) {
+        return { stashPushed: false, summary: "" };
+    }
+    // Warn the user before stashing
+    const warnMsg = `Working tree has uncommitted changes before milestone ${milestoneId} merge. Auto-stashing to allow clean merge (stash will be restored after merge).`;
+    notify(warnMsg, "warning");
+    // Push the stash
+    try {
+        execFileSync("git", ["stash", "push", "--include-untracked", "-m", "gsd-preflight-stash"], {
+            cwd: basePath,
+            stdio: ["ignore", "pipe", "pipe"],
+            encoding: "utf-8",
+            env: GIT_NO_PROMPT_ENV,
+        });
+        return {
+            stashPushed: true,
+            summary: `Stashed uncommitted changes before merge (milestone ${milestoneId}).`,
+        };
+    }
+    catch (err) {
+        // Stash failure is non-fatal — log and let the merge attempt proceed
+        const msg = `git stash push failed before merge of milestone ${milestoneId}: ${err instanceof Error ? err.message : String(err)}`;
+        logWarning("preflight", msg);
+        notify(`Auto-stash failed before milestone ${milestoneId} merge — proceeding anyway. ${msg}`, "warning");
+        return { stashPushed: false, summary: `stash-push-failed: ${msg}` };
+    }
+}
+/**
+ * Restore stashed changes after a milestone merge completes.
+ *
+ * Only called when preflightCleanRoot returned stashPushed=true.
+ * Any pop error (e.g. conflict) is logged and notified but does NOT throw —
+ * the merge already completed successfully.
+ */
+export function postflightPopStash(basePath, milestoneId, notify) {
+    try {
+        execFileSync("git", ["stash", "pop"], {
+            cwd: basePath,
+            stdio: ["ignore", "pipe", "pipe"],
+            encoding: "utf-8",
+            env: GIT_NO_PROMPT_ENV,
+        });
+        notify(`Restored stashed changes after milestone ${milestoneId} merge.`, "info");
+    }
+    catch (err) {
+        // Pop conflicts mean the merged code collides with the stashed changes.
+        // Log a warning — the user needs to resolve manually, but the merge succeeded.
+        const msg = `git stash pop failed after merge of milestone ${milestoneId}: ${err instanceof Error ? err.message : String(err)}. Run "git stash pop" manually to restore your changes.`;
+        logWarning("preflight", msg);
+        notify(msg, "warning");
+    }
+}