@muggleai/works 5.20.0-staging.135 → 5.20.0-staging.137

package/dist/{chunk-4M3P45WX.js → chunk-BA6EYJRB.js}

@@ -354,15 +354,37 @@ function parentIsAlive(pid) {
   }
 }
 
-// src/cli/pr-section/render.ts
+// src/cli/pr-section/constants.ts
 var DASHBOARD_URL_BASE = "https://www.muggle-ai.com/muggleTestV0/dashboard/projects";
 var DETAIL_IMAGE_WIDTH = 720;
-function countTests(report) {
-  const total = report.tests.length;
-  const passed = report.tests.filter((t) => t.status === "passed").length;
-  const failed = report.tests.filter((t) => t.status === "failed").length;
-  const inconclusive = report.tests.filter((t) => t.status === "inconclusive").length;
-  return { total, passed, failed, inconclusive };
+var STEP_LIST_ELISION_THRESHOLD = 10;
+var STEP_LIST_HEAD_COUNT = 6;
+var STEP_LIST_TAIL_COUNT = 3;
+
+// src/cli/pr-section/render.ts
+function resolveLatestRuns(tests) {
+  const byCaseId = /* @__PURE__ */ new Map();
+  const order = [];
+  for (const test of tests) {
+    const existing = byCaseId.get(test.testCaseId);
+    if (existing) {
+      existing.test = test;
+      existing.attempts += 1;
+      continue;
+    }
+    byCaseId.set(test.testCaseId, { test, attempts: 1 });
+    order.push(test.testCaseId);
+  }
+  return order.map((id) => byCaseId.get(id));
+}
+function countTests(latestRuns) {
+  const statuses = latestRuns.map((entry) => entry.test.status);
+  return {
+    total: statuses.length,
+    passed: statuses.filter((s) => s === "passed").length,
+    failed: statuses.filter((s) => s === "failed").length,
+    inconclusive: statuses.filter((s) => s === "inconclusive").length
+  };
 }
 function statusEmoji(test) {
   if (test.status === "passed") return "\u2705";
@@ -370,23 +392,30 @@ function statusEmoji(test) {
   return "\u26A0\uFE0F";
 }
 function computeVerdict(report) {
-  if (report.tests.length === 0) return "none";
-  if (report.tests.some((t) => t.status === "failed")) return "fail";
-  if (report.tests.some((t) => t.status === "inconclusive")) return "inconclusive";
+  const latestRuns = resolveLatestRuns(report.tests);
+  if (latestRuns.length === 0) return "none";
+  if (latestRuns.some((entry) => entry.test.status === "failed")) return "fail";
+  if (latestRuns.some((entry) => entry.test.status === "inconclusive")) return "inconclusive";
   return "pass";
 }
-function renderVerdictLine(verdict) {
+function verdictLabel(verdict) {
   switch (verdict) {
     case "pass":
-      return "**Verdict: \u2705 PASS**";
+      return "\u2705 PASS";
     case "fail":
-      return "**Verdict: \u274C FAIL**";
+      return "\u274C FAIL";
     case "inconclusive":
-      return "**Verdict: \u26A0\uFE0F INCONCLUSIVE**";
+      return "\u26A0\uFE0F INCONCLUSIVE";
     case "none":
       return null;
   }
 }
+function renderSummaryHeadline(verdict, counts) {
+  const { total, passed, failed, inconclusive } = counts;
+  const tally = `${total} ${total === 1 ? "test" : "tests"} ran \xB7 ${passed} passed / ${failed} failed / ${inconclusive} inconclusive`;
+  const label = verdictLabel(verdict);
+  return label ? `**${label}** \u2014 ${tally}` : `**${tally}**`;
+}
 function endingScreenshot(test) {
   if (test.steps.length === 0) {
     return null;
@@ -400,9 +429,13 @@ function endingScreenshot(test) {
   }
   return test.steps[test.steps.length - 1];
 }
+function failureOrdinal(test, failureStepIndex) {
+  const position = test.steps.findIndex((s) => s.stepIndex === failureStepIndex);
+  return position === -1 ? failureStepIndex : position + 1;
+}
 function defaultEndingCaption(test) {
   if (test.status === "failed") {
-    return `Failure at step ${test.failureStepIndex}`;
+    return `Failure at step ${failureOrdinal(test, test.failureStepIndex)}`;
   }
   if (test.status === "inconclusive") {
     return "Last frame before run was cut short";
@@ -431,88 +464,93 @@ function fullSizeImage(url, alt) {
 function safeInlineCode(s) {
   return s.replace(/`/g, "\u2018");
 }
-function buildTestNumbering(report) {
-  const map = /* @__PURE__ */ new Map();
-  report.tests.forEach((t, i) => {
-    map.set(t, i + 1);
-  });
-  return map;
-}
 function renderOverview(report) {
-  const { total, passed, failed, inconclusive } = countTests(report);
-  const verdict = computeVerdict(report);
-  const verdictLine = renderVerdictLine(verdict);
-  const lines = ["## E2E Acceptance Results", ""];
-  if (verdictLine) {
-    lines.push(verdictLine, "");
-  }
-  lines.push(`**${total} tests ran \u2014 ${passed} passed / ${failed} failed / ${inconclusive} inconclusive**`);
-  if (total === 0) {
+  const latestRuns = resolveLatestRuns(report.tests);
+  const counts = countTests(latestRuns);
+  const lines = [
+    "## E2E Acceptance Results",
+    "",
+    "### Summary",
+    "",
+    renderSummaryHeadline(computeVerdict(report), counts)
+  ];
+  if (counts.total === 0) {
     lines.push("", "_No tests were executed._");
     return lines.join("\n");
   }
-  lines.push("", "**Tests run:**");
-  const numbering = buildTestNumbering(report);
-  const anyGrouped = report.tests.some((t) => Boolean(t.useCaseName));
+  lines.push("");
+  const numbering = buildTestNumbering(latestRuns);
+  const anyGrouped = latestRuns.some((entry) => Boolean(entry.test.useCaseName));
   if (!anyGrouped) {
-    for (const t of report.tests) {
-      lines.push(`- **${numbering.get(t)}.** ${statusEmoji(t)} ${t.name}`);
+    for (const entry of latestRuns) {
+      lines.push(renderOverviewBullet(entry, numbering, ""));
     }
     return lines.join("\n");
   }
-  const groupOrder = [];
-  const groups = /* @__PURE__ */ new Map();
-  const flat = [];
-  const seenGroups = /* @__PURE__ */ new Set();
-  for (const t of report.tests) {
-    if (t.useCaseName) {
-      if (!groups.has(t.useCaseName)) {
-        groups.set(t.useCaseName, []);
-        groupOrder.push(t.useCaseName);
-      }
-      groups.get(t.useCaseName).push(t);
-      if (!seenGroups.has(t.useCaseName)) {
-        seenGroups.add(t.useCaseName);
-        flat.push({ type: "group", key: t.useCaseName });
-      }
-    } else {
-      flat.push({ type: "test", test: t });
-    }
-  }
-  for (const entry of flat) {
+  for (const entry of buildGroupedLayout(latestRuns)) {
     if (entry.type === "test") {
-      lines.push(`- **${numbering.get(entry.test)}.** ${statusEmoji(entry.test)} ${entry.test.name}`);
+      lines.push(renderOverviewBullet(entry.entry, numbering, ""));
     } else {
       lines.push(`- **${entry.key}**`);
-      for (const t of groups.get(entry.key)) {
-        lines.push(`  - **${numbering.get(t)}.** ${statusEmoji(t)} ${t.name}`);
+      for (const member of entry.members) {
+        lines.push(renderOverviewBullet(member, numbering, "  "));
       }
     }
   }
   return lines.join("\n");
 }
-function renderTestDetails(test, projectId, testNumber, dashboardBaseUrl = DASHBOARD_URL_BASE) {
-  const summary = renderSummaryLine(test, testNumber);
-  const frameBlock = renderEndingFrame(test);
-  const resultLines = renderResultSummary(test, projectId, dashboardBaseUrl);
+function buildTestNumbering(latestRuns) {
+  const map = /* @__PURE__ */ new Map();
+  latestRuns.forEach((entry, i) => {
+    map.set(entry.test.testCaseId, i + 1);
+  });
+  return map;
+}
+function renderOverviewBullet(entry, numbering, indent) {
+  const rerunNote = entry.attempts > 1 ? ` _(${entry.attempts} runs)_` : "";
+  return `${indent}- ${statusEmoji(entry.test)} **${numbering.get(entry.test.testCaseId)}.** ${entry.test.name}${rerunNote}`;
+}
+function buildGroupedLayout(latestRuns) {
+  const groups = /* @__PURE__ */ new Map();
+  const layout = [];
+  for (const entry of latestRuns) {
+    const useCaseName = entry.test.useCaseName;
+    if (!useCaseName) {
+      layout.push({ type: "test", entry });
+      continue;
+    }
+    const existing = groups.get(useCaseName);
+    if (existing) {
+      existing.push(entry);
+      continue;
+    }
+    const members = [entry];
+    groups.set(useCaseName, members);
+    layout.push({ type: "group", key: useCaseName, members });
+  }
+  return layout;
+}
+function renderTestDetails(entry, projectId, testNumber, dashboardBaseUrl = DASHBOARD_URL_BASE) {
+  const { test } = entry;
   const body = ["", "<br>", ""];
+  body.push(...renderResultSummary(entry, projectId, dashboardBaseUrl));
+  const stepList = renderStepList(test.steps);
+  if (stepList) {
+    body.push("", ...stepList);
+  }
+  const frameBlock = renderEndingFrame(test);
   if (frameBlock) {
-    body.push(...frameBlock, "");
+    body.push("", ...frameBlock);
   }
-  body.push(...resultLines);
   return `<details>
-<summary>${summary}</summary>
+<summary>${renderSummaryLine(test, testNumber)}</summary>
 ${body.join("\n")}
 
 </details>`;
 }
 function renderSummaryLine(test, testNumber) {
-  const base = `<b>${test.name}</b> ${statusEmoji(test)}`;
-  const head = "<i>\u25B6 click to expand</i> ";
-  if (test.description) {
-    return `<b>${testNumber}. </b>${head}${base} \u2014 ${test.description}`;
-  }
-  return `<b>${testNumber}. </b>${head}${base}`;
+  const base = `${statusEmoji(test)} <b>${testNumber}. ${test.name}</b>`;
+  return test.description ? `${base} \u2014 ${test.description}` : base;
 }
 function renderEndingFrame(test) {
   const frame = endingFrame(test);
@@ -525,22 +563,42 @@ function renderEndingFrame(test) {
     fullSizeImage(frame.url, test.name)
   ];
 }
-function renderResultSummary(test, projectId, dashboardBaseUrl) {
+function renderStepList(steps) {
+  if (steps.length === 0) {
+    return null;
+  }
+  const entry = (step, position) => `${position}. ${safeInlineCode(step.action)}`;
+  if (steps.length <= STEP_LIST_ELISION_THRESHOLD) {
+    return steps.map((step, i) => entry(step, i + 1));
+  }
+  const head = steps.slice(0, STEP_LIST_HEAD_COUNT).map((step, i) => entry(step, i + 1));
+  const tailStart = steps.length - STEP_LIST_TAIL_COUNT;
+  const tail = steps.slice(tailStart).map((step, i) => entry(step, tailStart + i + 1));
+  const hidden = tailStart - STEP_LIST_HEAD_COUNT;
+  return [...head, "", `_\u2026 ${hidden} more steps \u2026_`, "", ...tail];
+}
+function renderResultSummary(entry, projectId, dashboardBaseUrl) {
+  const { test } = entry;
   const dashboardUrl = `${dashboardBaseUrl}/${projectId}/scripts?modal=script-details&testCaseId=${encodeURIComponent(test.testCaseId)}`;
   const lines = [];
   if (test.status === "passed") {
     lines.push(`**Result:** \u2705 PASSED`);
   } else if (test.status === "failed") {
-    lines.push(`**Result:** \u274C FAILED at step ${test.failureStepIndex}`);
+    lines.push(`**Result:** \u274C FAILED at step ${failureOrdinal(test, test.failureStepIndex)}`);
     lines.push(`**Error:** \`${safeInlineCode(test.error)}\``);
   } else {
     lines.push(`**Result:** \u26A0\uFE0F INCONCLUSIVE`);
     lines.push(`**Reason:** \`${safeInlineCode(test.reason)}\``);
   }
-  lines.push(`**Steps:** ${test.steps.length}`);
-  lines.push(`<a href="${dashboardUrl}" target="_blank" rel="noopener noreferrer">View steps on Muggle AI \u2192</a>`);
+  if (entry.attempts > 1) {
+    lines.push(`**Runs:** ${entry.attempts} \u2014 showing the latest`);
+  }
+  lines.push(`**Steps:** ${test.steps.length} \xB7 <a href="${dashboardUrl}" target="_blank" rel="noopener noreferrer">View steps on Muggle AI \u2192</a>`);
   return lines;
 }
+function renderDetailBlocks(report, dashboardBaseUrl) {
+  return resolveLatestRuns(report.tests).map((entry, i) => renderTestDetails(entry, report.projectId, i + 1, dashboardBaseUrl ?? DASHBOARD_URL_BASE)).join("\n\n");
+}
 function renderBody(report, opts) {
   const overview = renderOverview(report);
   if (report.tests.length === 0) {
@@ -555,26 +613,28 @@ function renderBody(report, opts) {
       "_Full per-test details in the comment below \u2014 the PR description was too large to inline them._"
     ].join("\n");
   }
-  const detailBlocks = report.tests.map((t, i) => renderTestDetails(t, report.projectId, i + 1, opts.dashboardBaseUrl ?? DASHBOARD_URL_BASE));
   return [
     overview,
     "",
     "---",
     "",
-    detailBlocks.join("\n\n")
+    "### Test details",
+    "",
+    renderDetailBlocks(report, opts.dashboardBaseUrl)
   ].join("\n");
 }
 function renderComment(report, opts = {}) {
   if (report.tests.length === 0) {
     return "";
   }
-  const detailBlocks = report.tests.map((t, i) => renderTestDetails(t, report.projectId, i + 1, opts.dashboardBaseUrl ?? DASHBOARD_URL_BASE));
   return [
     "## E2E acceptance evidence (overflow)",
     "",
     "_This comment was posted because the full per-test details did not fit in the PR description._",
     "",
-    detailBlocks.join("\n\n")
+    "### Test details",
+    "",
+    renderDetailBlocks(report, opts.dashboardBaseUrl)
   ].join("\n");
 }
 

package/dist/cli.js

@@ -1,5 +1,5 @@
 #!/usr/bin/env node
-import { runCli } from './chunk-4M3P45WX.js';
+import { runCli } from './chunk-BA6EYJRB.js';
 import './chunk-ZA5KHVX5.js';
 
 // src/cli/main.ts

package/dist/index.js

@@ -1,2 +1,2 @@
-export { src_exports as commands, createUnifiedMcpServer, server_exports as server } from './chunk-4M3P45WX.js';
+export { src_exports as commands, createUnifiedMcpServer, server_exports as server } from './chunk-BA6EYJRB.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-ZA5KHVX5.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": "5.20.0-staging.135",
+  "version": "5.20.0-staging.137",
   "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": "5.20.0-staging.135",
+  "version": "5.20.0-staging.137",
   "author": {
     "name": "Muggle AI",
     "email": "support@muggle-ai.com"

package/dist/release-manifest.json

@@ -1,7 +1,7 @@
 {
   "release": "5.18.0",
-  "buildId": "run-135-1",
-  "commitSha": "605057934e0d2ad35b957e511584bfac7cd7d400",
-  "buildTime": "2026-09-24T17:28:38Z",
+  "buildId": "run-137-1",
+  "commitSha": "0ac93fb98732cc7c350e2269b19b552438f0fced",
+  "buildTime": "2026-09-24T18:14:39Z",
   "serviceName": "muggle-ai-works-mcp"
 }

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@muggleai/works",
   "mcpName": "io.github.multiplex-ai/muggle",
-  "version": "5.20.0-staging.135",
+  "version": "5.20.0-staging.137",
   "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",

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": "5.20.0-staging.135",
+  "version": "5.20.0-staging.137",
   "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": "5.20.0-staging.135",
+  "version": "5.20.0-staging.137",
   "author": {
     "name": "Muggle AI",
     "email": "support@muggle-ai.com"