@muggleai/works 4.11.0 → 4.11.2

package/dist/{chunk-Q7WNEMN4.js → chunk-7WRAMBOS.js}

@@ -1,4 +1,4 @@
-import { __export, getLogger, getConfig, createChildLogger, buildElectronAppReleaseAssetUrl, getAuthService, hasApiKey, getElectronAppVersion, getElectronAppDir, getPlatformKey, isFirstRun, writePreferences, DEFAULT_PREFERENCES, getDataDir, PREFERENCES_FILE_NAME, isElectronAppInstalled, getElectronAppChecksums, getChecksumForPlatform, verifyFileChecksum, calculateFileChecksum, initTelemetry, Surface, ServiceName, track, EventName, getQaTools, getLocalQaTools, performLogout, performLogin, toolRequiresAuth, getCallerCredentials, getBundledElectronAppVersion, getElectronAppVersionSource, getCredentialsFilePath, buildElectronAppChecksumsUrl, __require } from './chunk-ZKRHFTIC.js';
+import { __export, getLogger, getConfig, createChildLogger, buildElectronAppReleaseAssetUrl, getAuthService, hasApiKey, getElectronAppVersion, getElectronAppDir, getPlatformKey, isFirstRun, writePreferences, DEFAULT_PREFERENCES, getDataDir, PREFERENCES_FILE_NAME, isElectronAppInstalled, getElectronAppChecksums, getChecksumForPlatform, verifyFileChecksum, calculateFileChecksum, initTelemetry, Surface, ServiceName, track, EventName, getQaTools, getLocalQaTools, performLogout, performLogin, toolRequiresAuth, getCallerCredentials, hasShownDisclosure, getDisclosureCopy, markDisclosureShown, getBundledElectronAppVersion, getElectronAppVersionSource, getCredentialsFilePath, buildElectronAppChecksumsUrl, __require } from './chunk-AYWLYB3F.js';
 import { Server } from '@modelcontextprotocol/sdk/server/index.js';
 import { ListToolsRequestSchema, CallToolRequestSchema, ListResourcesRequestSchema, ReadResourceRequestSchema } from '@modelcontextprotocol/sdk/types.js';
 import { v4 } from 'uuid';
@@ -731,7 +731,7 @@ async function resolveGsScreenshotUrls(report, opts) {
     if (gsUrls.length === 0) {
       return report;
     }
-    const mcps = await import('./src-2G7ENRGI.js');
+    const mcps = await import('./src-NJUUHKPW.js');
     const credentials = await mcps.getCallerCredentialsAsync();
     if (!credentials.bearerToken && !credentials.apiKey) {
       stderrWrite(
@@ -1638,6 +1638,15 @@ async function statusCommand() {
 }
 
 // src/cli/serve.ts
+var APPLICATIONINSIGHTS_CONNECTION_STRING = "InstrumentationKey=0e28df42-b98e-4089-b3b5-6decefb99447;IngestionEndpoint=https://westus3-1.in.applicationinsights.azure.com/;LiveEndpoint=https://westus3.livediagnostics.monitor.azure.com/;ApplicationId=b64a932f-85ec-40cd-a74b-e181a3660f6c";
+function showDisclosureIfNeeded() {
+  try {
+    if (hasShownDisclosure()) return;
+    process.stderr.write(getDisclosureCopy() + "\n");
+    markDisclosureShown();
+  } catch {
+  }
+}
 var logger5 = getLogger();
 async function serveCommand(options) {
   const config = getConfig();
@@ -1650,7 +1659,12 @@ async function serveCommand(options) {
     transport: "stdio"
   });
   try {
-    initTelemetry({ serviceName: ServiceName.MuggleMcp, surface: Surface.McpLocal });
+    showDisclosureIfNeeded();
+    initTelemetry({
+      serviceName: ServiceName.MuggleMcp,
+      surface: Surface.McpLocal,
+      connectionString: APPLICATIONINSIGHTS_CONNECTION_STRING
+    });
     track({ name: EventName.SystemStartup, props: { serviceName: ServiceName.MuggleMcp } });
   } catch (err) {
     logger5.warn("Telemetry init skipped", {

package/dist/{chunk-ZKRHFTIC.js → chunk-AYWLYB3F.js}

@@ -94,7 +94,7 @@ __export(src_exports, {
   toolRequiresAuth: () => toolRequiresAuth,
   validatePreference: () => validatePreference,
   verifyFileChecksum: () => verifyFileChecksum,
-  writePreferences: () => writePreferences
+  writePreferences: () => writePreferences2
 });
 var DATA_DIR_NAME = ".muggle-ai";
 function getDataDir() {
@@ -3198,6 +3198,19 @@ function readPreferences() {
     return {};
   }
 }
+function writePreferences(prefs) {
+  const filePath = prefsFilePath();
+  mkdirSync(dirname(filePath), { recursive: true });
+  writeFileSync(filePath, JSON.stringify(prefs, null, 2) + "\n", "utf8");
+}
+function updatePreference(key, value) {
+  const prefs = readPreferences();
+  prefs[key] = value;
+  writePreferences(prefs);
+}
+function preferencesFileExists() {
+  return existsSync(prefsFilePath());
+}
 
 function isTelemetryEnabled() {
   if (process.env[ENV_TELEMETRY_DISABLED] === "1")
@@ -3223,6 +3236,21 @@ function track(event) {
   }
 }
 
+var DISCLOSURE_COPY = `Muggle Test collects anonymous usage telemetry \u2014 which features you use, how long they take, and whether they succeed \u2014 to help us improve the product. We attach an opaque install ID, your account ID (when signed in), and your IP. We do not collect file contents, OS username, hostname, or file paths. To opt out, set telemetryEnabled to false in ~/.muggle-ai/preferences.json or run \`muggle preferences set telemetryEnabled false\`.`;
+
+function getDisclosureCopy() {
+  return DISCLOSURE_COPY;
+}
+function hasShownDisclosure() {
+  if (!preferencesFileExists())
+    return false;
+  const prefs = readPreferences();
+  return typeof prefs.disclosureShownAt === "string" && prefs.disclosureShownAt.length > 0;
+}
+function markDisclosureShown(now = /* @__PURE__ */ new Date()) {
+  updatePreference("disclosureShownAt", now.toISOString());
+}
+
 var Trigger;
 (function(Trigger2) {
   Trigger2["UserSlash"] = "user-slash";
@@ -6018,7 +6046,7 @@ function resolvePreferences(dataDirOverride2, cwd) {
   const project = readProjectPreferences(cwd);
   return { ...global, ...project };
 }
-function writePreferences(prefs, scope, dataDirOverride2, cwd) {
+function writePreferences2(prefs, scope, dataDirOverride2, cwd) {
   const dir = scope === "project" && cwd ? path2.join(cwd, PREFERENCES_PROJECT_DIR_NAME) : dataDirOverride2 ?? getDataDir();
   fs5.mkdirSync(dir, { recursive: true });
   const filePath = path2.join(dir, PREFERENCES_FILE_NAME);
@@ -6584,7 +6612,7 @@ var preferencesSetTool = {
     const logger6 = createChildLogger2(ctx.correlationId);
     logger6.info("Executing muggle-local-preferences-set");
     const input = PreferencesSetInputSchema.parse(ctx.input);
-    writePreferences(
+    writePreferences2(
       { [input.key]: input.value },
       input.scope,
       void 0,
@@ -7056,4 +7084,4 @@ var plugins_exports = {};
 // packages/mcps/src/mcp/agents/index.ts
 var agents_exports = {};
 
-export { DEFAULT_PREFERENCES, EventName, PREFERENCES_FILE_NAME, PREFERENCES_PROJECT_DIR_NAME, PREFERENCES_SCHEMA, PREFERENCES_VERSION, PREFERENCE_ALLOWED_VALUES, PreferenceKey, PreferenceValue, ServiceName, Surface, __export, __require, buildElectronAppChecksumsUrl, buildElectronAppReleaseAssetUrl, buildElectronAppReleaseTag, calculateFileChecksum, createApiKeyWithToken, createChildLogger, deleteApiKeyData, deleteCredentials, e2e_exports2 as e2e_exports, formatPreferencesOneLiner, getApiKey, getApiKeyFilePath, getAuthService, getBundledElectronAppVersion, getCallerCredentials, getCallerCredentialsAsync, getChecksumForPlatform, getConfig, getCredentialsFilePath, getDataDir2 as getDataDir, getDownloadBaseUrl, getElectronAppChecksums, getElectronAppDir, getElectronAppVersion, getElectronAppVersionSource, getLocalQaTools, getLogger, getPlatformKey, getQaTools, getValidApiKeyData, getValidCredentials, hasApiKey, initTelemetry, isElectronAppInstalled, isFirstRun, loadApiKeyData, loadCredentials, local_exports2 as local_exports, mcp_exports, openBrowserUrl, performLogin, performLogout, pollDeviceCode, readGlobalPreferences, readProjectPreferences, resetConfig, resetLogger, resetPreference, resolveElectronAppPathOrNull, resolvePreferences, saveApiKey, saveApiKeyData, saveCredentials, src_exports, startDeviceCodeFlow, toolRequiresAuth, track, validatePreference, verifyFileChecksum, writePreferences };
+export { DEFAULT_PREFERENCES, EventName, PREFERENCES_FILE_NAME, PREFERENCES_PROJECT_DIR_NAME, PREFERENCES_SCHEMA, PREFERENCES_VERSION, PREFERENCE_ALLOWED_VALUES, PreferenceKey, PreferenceValue, ServiceName, Surface, __export, __require, buildElectronAppChecksumsUrl, buildElectronAppReleaseAssetUrl, buildElectronAppReleaseTag, calculateFileChecksum, createApiKeyWithToken, createChildLogger, deleteApiKeyData, deleteCredentials, e2e_exports2 as e2e_exports, formatPreferencesOneLiner, getApiKey, getApiKeyFilePath, getAuthService, getBundledElectronAppVersion, getCallerCredentials, getCallerCredentialsAsync, getChecksumForPlatform, getConfig, getCredentialsFilePath, getDataDir2 as getDataDir, getDisclosureCopy, getDownloadBaseUrl, getElectronAppChecksums, getElectronAppDir, getElectronAppVersion, getElectronAppVersionSource, getLocalQaTools, getLogger, getPlatformKey, getQaTools, getValidApiKeyData, getValidCredentials, hasApiKey, hasShownDisclosure, initTelemetry, isElectronAppInstalled, isFirstRun, loadApiKeyData, loadCredentials, local_exports2 as local_exports, markDisclosureShown, mcp_exports, openBrowserUrl, performLogin, performLogout, pollDeviceCode, readGlobalPreferences, readProjectPreferences, resetConfig, resetLogger, resetPreference, resolveElectronAppPathOrNull, resolvePreferences, saveApiKey, saveApiKeyData, saveCredentials, src_exports, startDeviceCodeFlow, toolRequiresAuth, track, validatePreference, verifyFileChecksum, writePreferences2 as writePreferences };

package/dist/cli.js

@@ -1,6 +1,6 @@
 #!/usr/bin/env node
-import { runCli } from './chunk-Q7WNEMN4.js';
-import './chunk-ZKRHFTIC.js';
+import { runCli } from './chunk-7WRAMBOS.js';
+import './chunk-AYWLYB3F.js';
 
 // src/cli/main.ts
 runCli().catch((error) => {

package/dist/index.js

@@ -1,2 +1,2 @@
-export { src_exports as commands, createUnifiedMcpServer, server_exports as server } from './chunk-Q7WNEMN4.js';
-export { createChildLogger, e2e_exports as e2e, getConfig, getLocalQaTools, getLogger, getQaTools, local_exports as localQa, mcp_exports as mcp, e2e_exports as qa, src_exports as shared } from './chunk-ZKRHFTIC.js';
+export { src_exports as commands, createUnifiedMcpServer, server_exports as server } from './chunk-7WRAMBOS.js';
+export { createChildLogger, e2e_exports as e2e, getConfig, getLocalQaTools, getLogger, getQaTools, local_exports as localQa, mcp_exports as mcp, e2e_exports as qa, src_exports as shared } from './chunk-AYWLYB3F.js';

package/dist/plugin/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "muggle",
   "description": "Run real-browser end-to-end (E2E) acceptance tests on your web app from any AI coding agent. Generate test scripts from plain English, replay them on localhost, capture screenshots, and validate user flows like signup, checkout, and dashboards. Works across Claude Code, Cursor, Codex, and Windsurf.",
-  "version": "4.11.0",
+  "version": "4.11.1",
   "author": {
     "name": "Muggle AI",
     "email": "support@muggle-ai.com"

package/dist/plugin/.cursor-plugin/plugin.json

@@ -2,7 +2,7 @@
   "name": "muggle",
   "displayName": "Muggle AI",
   "description": "Ship quality products with AI-powered end-to-end (E2E) acceptance testing that validates your web app like a real user — from Claude Code and Cursor to PR.",
-  "version": "4.11.0",
+  "version": "4.11.1",
   "author": {
     "name": "Muggle AI",
     "email": "support@muggle-ai.com"

package/dist/plugin/skills/_shared/pr-followup-helpers.md

@@ -101,18 +101,19 @@ fix(ci): lint — remove unused import
 
 ## Classify
 
-Classify the **review as a unit**, not individual comments.
+Classify the **review as a unit** — but reply per line comment (threaded), not per review.
 
 | Class | Signal | Action |
 | :---- | :----- | :----- |
-| **actionable** | Review names at least one concrete change or asks an answerable question. Soft phrasing counts when there's a concrete referent. | Treat as amended requirements; run the caller's implementation cycle; reply with one summary referencing the new SHA. |
+| **actionable** | Review names at least one concrete change or asks an answerable question. Soft phrasing counts when there's a concrete referent. | Treat as amended requirements; run **one** implementation cycle for the whole review; reply **threaded per line comment** referencing the new SHA (top-level only when the review is body-only). |
 | **ambiguous** | No actionable signal — pure vibes, contradictory, or depends on knowledge the loop can't access. | Escalate once with two interpretations; pause the PR. |
 
 Default to **actionable**. CI catches wrong attempts; reviewers correct on the next round. Escalation is a round-trip with an absent user — reserve it.
 
-Reply summary shape:
+Reply shape (all replies for one review reference the same SHA):
 
-- **actionable**: `Addressed review <review_id> in <sha> — cycle ran clean (or: with <N> failures, see walkthrough).`
+- **threaded** (default): `Done in <sha> — <attribution>. (Review #<review_id>, cycle <status>.)`
+- **top-level** (fallback, body-only reviews): `Re: review #<review_id> — addressed in <sha>, cycle <status>.`
 - **ambiguous**: no bot reply.
 
 ### Worked examples — Actionable reviews

package/dist/plugin/skills/muggle-pr-followup/contract.md

@@ -86,10 +86,7 @@ When the review is actionable:
 2. **Amend `requirements.md`** in the session dir with a new `## Amendment — review <review_id> by <login> (<timestamp>)` section pasting the review body and each comment (with `<file>:<line>` context).
 3. **Invoke the implementation cycle** declared in the caller's `cycle.json`. Iterate the `steps[]` in order. Each step is either a markdown file to follow, a skill to invoke, or a shell command (per the `cycle.json` schema in SKILL.md). When a step fails, the cycle returns `failed: <step-name>`; the loop escalates per Step 8 with the failure as the reason.
 4. **Push** via `cycle.json`'s `pushHandler`. Set `last_seen.last_pushed_sha` to the new HEAD.
-5. **Reply** with one summary via `gh pr comment <n>`:
-   ```
-   Addressed review <review_id> in <sha> — cycle ran clean (or: with <N> failures, see walkthrough).
-   ```
+5. **Reply** per [helpers § Reply routing](../_shared/pr-followup-helpers.md#reply-routing) and [§ Classify](../_shared/pr-followup-helpers.md#classify) (reply shape). For each line comment in the review, derive `<attribution>` from `git diff <last_pushed_sha>..HEAD -- <comment.path>` near `comment.line` ±5 (fall back to `addressed indirectly — see walkthrough` if empty). `<status>` = `ran clean` or `had <N> failures, see walkthrough`. If the review is body-only (no line comments), post the top-level fallback shape; if both body and line comments, threaded replies cover it — no top-level.
 6. **Resume polling**: clear `cycling: true`, increment `cycles_completed`, advance `last_seen.reviewId` past this review.
 7. Emit per-cycle telemetry.
 
@@ -133,9 +130,10 @@ Emit one tick event per `muggle-local-telemetry-skill-emit`. Exit the turn.
 
 ## Reply routing
 
-- **Summary reply on a review**: `gh pr comment <number> --body "..."` referencing the review id and the new SHA. There's no "reply to a review" endpoint.
-- **Reply to a specific line comment** (optional): `POST /repos/{owner}/{repo}/pulls/{n}/comments/{comment_id}/replies`.
-- **Never post the same summary twice** — `last_seen.reviewId` is the only re-entry guard.
+- **Threaded reply per line comment** (default): `POST /repos/{owner}/{repo}/pulls/{n}/comments/{comment_id}/replies`. Use for every line comment in the review so each thread can be resolved in GitHub's UI.
+- **Top-level summary on a body-only review** (fallback): `gh pr comment <number> --body "..."` referencing the review id and the new SHA. Used only when the review has body content and zero line comments — GitHub has no "reply to a review body" endpoint.
+- **Never post the same reply twice** — `last_seen.reviewId` is the only re-entry guard.
+- **Never post a top-level summary alongside threaded replies** — duplication pollutes the Conversation tab.
 
 ## Telemetry
 
@@ -187,5 +185,6 @@ This stage produces no console output beyond:
 - [ ] `followup.log` has at minimum a heartbeat or per-review line for this tick.
 - [ ] Telemetry events emitted (per-cycle when applicable + per-tick).
 - [ ] If pushed, `last_pushed_sha` is set and `cycles_completed` incremented.
+- [ ] If actionable, one threaded reply posted per line comment (or one top-level reply for body-only reviews) — never both, never zero.
 - [ ] If escalated, `escalated_review_ids` contains the review id.
 - [ ] If terminal, the loop is NOT continued.

package/dist/release-manifest.json

@@ -1,7 +1,7 @@
 {
-  "release": "4.11.0",
-  "buildId": "run-33-1",
-  "commitSha": "f704789e0836e91f5e24f364d0d8eccf1d1a5439",
-  "buildTime": "2026-05-15T07:07:24Z",
+  "release": "4.11.1",
+  "buildId": "run-35-1",
+  "commitSha": "91afe33fb96f46115bd399141f165832253b4809",
+  "buildTime": "2026-05-19T23:10:43Z",
   "serviceName": "muggle-ai-works-mcp"
 }

package/dist/{src-2G7ENRGI.js → src-NJUUHKPW.js}

@@ -1 +1 @@
-export { DEFAULT_PREFERENCES, PREFERENCES_FILE_NAME, PREFERENCES_PROJECT_DIR_NAME, PREFERENCES_SCHEMA, PREFERENCES_VERSION, PREFERENCE_ALLOWED_VALUES, PreferenceKey, PreferenceValue, buildElectronAppChecksumsUrl, buildElectronAppReleaseAssetUrl, buildElectronAppReleaseTag, calculateFileChecksum, createApiKeyWithToken, createChildLogger, deleteApiKeyData, deleteCredentials, e2e_exports as e2e, formatPreferencesOneLiner, getApiKey, getApiKeyFilePath, getAuthService, getBundledElectronAppVersion, getCallerCredentials, getCallerCredentialsAsync, getChecksumForPlatform, getConfig, getCredentialsFilePath, getDataDir, getDownloadBaseUrl, getElectronAppChecksums, getElectronAppDir, getElectronAppVersion, getElectronAppVersionSource, getLocalQaTools, getLogger, getPlatformKey, getQaTools, getValidApiKeyData, getValidCredentials, hasApiKey, isElectronAppInstalled, isFirstRun, loadApiKeyData, loadCredentials, local_exports as localQa, mcp_exports as mcp, openBrowserUrl, performLogin, performLogout, pollDeviceCode, e2e_exports as qa, readGlobalPreferences, readProjectPreferences, resetConfig, resetLogger, resetPreference, resolveElectronAppPathOrNull, resolvePreferences, saveApiKey, saveApiKeyData, saveCredentials, startDeviceCodeFlow, toolRequiresAuth, validatePreference, verifyFileChecksum, writePreferences } from './chunk-ZKRHFTIC.js';
+export { DEFAULT_PREFERENCES, PREFERENCES_FILE_NAME, PREFERENCES_PROJECT_DIR_NAME, PREFERENCES_SCHEMA, PREFERENCES_VERSION, PREFERENCE_ALLOWED_VALUES, PreferenceKey, PreferenceValue, buildElectronAppChecksumsUrl, buildElectronAppReleaseAssetUrl, buildElectronAppReleaseTag, calculateFileChecksum, createApiKeyWithToken, createChildLogger, deleteApiKeyData, deleteCredentials, e2e_exports as e2e, formatPreferencesOneLiner, getApiKey, getApiKeyFilePath, getAuthService, getBundledElectronAppVersion, getCallerCredentials, getCallerCredentialsAsync, getChecksumForPlatform, getConfig, getCredentialsFilePath, getDataDir, getDownloadBaseUrl, getElectronAppChecksums, getElectronAppDir, getElectronAppVersion, getElectronAppVersionSource, getLocalQaTools, getLogger, getPlatformKey, getQaTools, getValidApiKeyData, getValidCredentials, hasApiKey, isElectronAppInstalled, isFirstRun, loadApiKeyData, loadCredentials, local_exports as localQa, mcp_exports as mcp, openBrowserUrl, performLogin, performLogout, pollDeviceCode, e2e_exports as qa, readGlobalPreferences, readProjectPreferences, resetConfig, resetLogger, resetPreference, resolveElectronAppPathOrNull, resolvePreferences, saveApiKey, saveApiKeyData, saveCredentials, startDeviceCodeFlow, toolRequiresAuth, validatePreference, verifyFileChecksum, writePreferences } from './chunk-AYWLYB3F.js';

package/package.json

@@ -1,7 +1,7 @@
 {
     "name": "@muggleai/works",
     "mcpName": "io.github.multiplex-ai/muggle",
-    "version": "4.11.0",
+    "version": "4.11.2",
     "description": "Ship quality products with AI-powered E2E acceptance testing that validates your web app like a real user — from Claude Code and Cursor to PR.",
     "type": "module",
     "main": "dist/index.js",
@@ -41,14 +41,14 @@
         "test:watch": "vitest"
     },
     "muggleConfig": {
-        "electronAppVersion": "1.0.84",
+        "electronAppVersion": "1.0.87",
         "downloadBaseUrl": "https://github.com/multiplex-ai/muggle-ai-works/releases/download",
         "runtimeTargetDefault": "production",
         "checksums": {
-            "darwin-arm64": "da978efea8e3c8633f96c5245b932f98988eb60dc94cced62cac20f4bdefbb20",
-            "darwin-x64": "9aa611d446494ab1d48674813cead2d669f8aea6869efea6f4f2beb13d18e4ac",
-            "win32-x64": "5220df3c394f2406201955f04e78d8faf764ea565fc59611484c8bbaad24635e",
-            "linux-x64": "5c519a473800b2726eef7a55c50d1d2c2e0ff78ab2d860dc869e2cd42fa59d74"
+            "darwin-arm64": "df04c98269e22ae427bb863956215b3f20cbf2ff0b51c037647d6e734a7b7502",
+            "darwin-x64": "e9f5b8a61f4a9f47743aaabb27d6e4c462dca5aa922ba66ae04f0a0133b8a7eb",
+            "win32-x64": "75e552672757d280259637bdcbe0aae0eecc2417ed9a6e5bd49f42d3466f66cf",
+            "linux-x64": "2c8b64cf549f9a6682db959b17c5f04578d632d5bcdba1ac9efbb45d6a097802"
         }
     },
     "dependencies": {

package/plugin/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "muggle",
   "description": "Run real-browser end-to-end (E2E) acceptance tests on your web app from any AI coding agent. Generate test scripts from plain English, replay them on localhost, capture screenshots, and validate user flows like signup, checkout, and dashboards. Works across Claude Code, Cursor, Codex, and Windsurf.",
-  "version": "4.11.0",
+  "version": "4.11.1",
   "author": {
     "name": "Muggle AI",
     "email": "support@muggle-ai.com"

package/plugin/.cursor-plugin/plugin.json

@@ -2,7 +2,7 @@
   "name": "muggle",
   "displayName": "Muggle AI",
   "description": "Ship quality products with AI-powered end-to-end (E2E) acceptance testing that validates your web app like a real user — from Claude Code and Cursor to PR.",
-  "version": "4.11.0",
+  "version": "4.11.1",
   "author": {
     "name": "Muggle AI",
     "email": "support@muggle-ai.com"

package/plugin/skills/_shared/pr-followup-helpers.md

@@ -101,18 +101,19 @@ fix(ci): lint — remove unused import
 
 ## Classify
 
-Classify the **review as a unit**, not individual comments.
+Classify the **review as a unit** — but reply per line comment (threaded), not per review.
 
 | Class | Signal | Action |
 | :---- | :----- | :----- |
-| **actionable** | Review names at least one concrete change or asks an answerable question. Soft phrasing counts when there's a concrete referent. | Treat as amended requirements; run the caller's implementation cycle; reply with one summary referencing the new SHA. |
+| **actionable** | Review names at least one concrete change or asks an answerable question. Soft phrasing counts when there's a concrete referent. | Treat as amended requirements; run **one** implementation cycle for the whole review; reply **threaded per line comment** referencing the new SHA (top-level only when the review is body-only). |
 | **ambiguous** | No actionable signal — pure vibes, contradictory, or depends on knowledge the loop can't access. | Escalate once with two interpretations; pause the PR. |
 
 Default to **actionable**. CI catches wrong attempts; reviewers correct on the next round. Escalation is a round-trip with an absent user — reserve it.
 
-Reply summary shape:
+Reply shape (all replies for one review reference the same SHA):
 
-- **actionable**: `Addressed review <review_id> in <sha> — cycle ran clean (or: with <N> failures, see walkthrough).`
+- **threaded** (default): `Done in <sha> — <attribution>. (Review #<review_id>, cycle <status>.)`
+- **top-level** (fallback, body-only reviews): `Re: review #<review_id> — addressed in <sha>, cycle <status>.`
 - **ambiguous**: no bot reply.
 
 ### Worked examples — Actionable reviews

package/plugin/skills/muggle-pr-followup/contract.md

@@ -86,10 +86,7 @@ When the review is actionable:
 2. **Amend `requirements.md`** in the session dir with a new `## Amendment — review <review_id> by <login> (<timestamp>)` section pasting the review body and each comment (with `<file>:<line>` context).
 3. **Invoke the implementation cycle** declared in the caller's `cycle.json`. Iterate the `steps[]` in order. Each step is either a markdown file to follow, a skill to invoke, or a shell command (per the `cycle.json` schema in SKILL.md). When a step fails, the cycle returns `failed: <step-name>`; the loop escalates per Step 8 with the failure as the reason.
 4. **Push** via `cycle.json`'s `pushHandler`. Set `last_seen.last_pushed_sha` to the new HEAD.
-5. **Reply** with one summary via `gh pr comment <n>`:
-   ```
-   Addressed review <review_id> in <sha> — cycle ran clean (or: with <N> failures, see walkthrough).
-   ```
+5. **Reply** per [helpers § Reply routing](../_shared/pr-followup-helpers.md#reply-routing) and [§ Classify](../_shared/pr-followup-helpers.md#classify) (reply shape). For each line comment in the review, derive `<attribution>` from `git diff <last_pushed_sha>..HEAD -- <comment.path>` near `comment.line` ±5 (fall back to `addressed indirectly — see walkthrough` if empty). `<status>` = `ran clean` or `had <N> failures, see walkthrough`. If the review is body-only (no line comments), post the top-level fallback shape; if both body and line comments, threaded replies cover it — no top-level.
 6. **Resume polling**: clear `cycling: true`, increment `cycles_completed`, advance `last_seen.reviewId` past this review.
 7. Emit per-cycle telemetry.
 
@@ -133,9 +130,10 @@ Emit one tick event per `muggle-local-telemetry-skill-emit`. Exit the turn.
 
 ## Reply routing
 
-- **Summary reply on a review**: `gh pr comment <number> --body "..."` referencing the review id and the new SHA. There's no "reply to a review" endpoint.
-- **Reply to a specific line comment** (optional): `POST /repos/{owner}/{repo}/pulls/{n}/comments/{comment_id}/replies`.
-- **Never post the same summary twice** — `last_seen.reviewId` is the only re-entry guard.
+- **Threaded reply per line comment** (default): `POST /repos/{owner}/{repo}/pulls/{n}/comments/{comment_id}/replies`. Use for every line comment in the review so each thread can be resolved in GitHub's UI.
+- **Top-level summary on a body-only review** (fallback): `gh pr comment <number> --body "..."` referencing the review id and the new SHA. Used only when the review has body content and zero line comments — GitHub has no "reply to a review body" endpoint.
+- **Never post the same reply twice** — `last_seen.reviewId` is the only re-entry guard.
+- **Never post a top-level summary alongside threaded replies** — duplication pollutes the Conversation tab.
 
 ## Telemetry
 
@@ -187,5 +185,6 @@ This stage produces no console output beyond:
 - [ ] `followup.log` has at minimum a heartbeat or per-review line for this tick.
 - [ ] Telemetry events emitted (per-cycle when applicable + per-tick).
 - [ ] If pushed, `last_pushed_sha` is set and `cycles_completed` incremented.
+- [ ] If actionable, one threaded reply posted per line comment (or one top-level reply for body-only reviews) — never both, never zero.
 - [ ] If escalated, `escalated_review_ids` contains the review id.
 - [ ] If terminal, the loop is NOT continued.