fix: address PR review comments for 0416-simplify_dev_setup

* fix(ci): add PR comment with test summary and block on failure

Add a `comment` job to the test workflow that parses JUnit XML artifacts
and posts a sticky PR comment showing pass/fail counts per suite, with
failed test names listed in a collapsible section and a link to the run.

Guards against fork PRs (read-only token) and stale overlapping runs
(skips comment if PR head has moved past our SHA).

* fix(ci): use payload SHA for staleness check, handle missing artifacts

- Replace context.sha (merge commit SHA) with
  context.payload.pull_request.head.sha so the staleness guard
  compares the correct values and the comment actually gets posted
- Add continue-on-error to download-artifact so cancelled runs
  gracefully fall through to the "no test results" message

* fix(ci): show warning icon for zero-test suites instead of failure

.github/workflows/test.yml

@@ -136,3 +136,124 @@ jobs:
             echo "See the uploaded \`junit-${{ matrix.suite }}\` artifact for details." >> "$GITHUB_STEP_SUMMARY"
             exit 1
           fi
+
+  comment:
+    name: PR test summary
+    needs: test
+    if: >-
+      always()
+      && github.event_name == 'pull_request'
+      && github.event.pull_request.head.repo.full_name == github.repository
+    runs-on: ubuntu-latest
+    permissions:
+      pull-requests: write
+      actions: read
+    steps:
+      - name: Download JUnit artifacts
+        uses: actions/download-artifact@v4
+        continue-on-error: true
+        with:
+          path: junit
+          pattern: junit-*
+
+      - name: Build comment body
+        run: |
+          python3 <<'PY'
+          import glob, os, xml.etree.ElementTree as ET
+
+          run_url = f"{os.environ['GITHUB_SERVER_URL']}/{os.environ['GITHUB_REPOSITORY']}/actions/runs/{os.environ['GITHUB_RUN_ID']}"
+          marker = "<!-- browseros-agent-tests-summary -->"
+
+          suites = []
+          failed_cases = []
+          total_tests = total_failed = total_skipped = 0
+
+          for xml_path in sorted(glob.glob("junit/junit-*/*.xml")):
+              suite_name = os.path.basename(os.path.dirname(xml_path)).removeprefix("junit-")
+              try:
+                  root = ET.parse(xml_path).getroot()
+              except ET.ParseError:
+                  suites.append({"name": suite_name, "passed": 0, "failed": 1, "skipped": 0, "total": 1})
+                  total_tests += 1
+                  total_failed += 1
+                  failed_cases.append((suite_name, "(could not parse junit XML)"))
+                  continue
+
+              testsuites = root.findall("testsuite") if root.tag == "testsuites" else [root]
+              s_tests = s_fail = s_err = s_skip = 0
+              for ts in testsuites:
+                  s_tests += int(ts.get("tests") or 0)
+                  s_fail += int(ts.get("failures") or 0)
+                  s_err += int(ts.get("errors") or 0)
+                  s_skip += int(ts.get("skipped") or 0)
+                  for tc in ts.iter("testcase"):
+                      if tc.find("failure") is not None or tc.find("error") is not None:
+                          cls = tc.get("classname") or ""
+                          name = tc.get("name") or "(unnamed)"
+                          label = f"{cls} > {name}" if cls else name
+                          failed_cases.append((suite_name, label))
+
+              s_failed = s_fail + s_err
+              s_passed = max(s_tests - s_failed - s_skip, 0)
+              suites.append({"name": suite_name, "passed": s_passed, "failed": s_failed, "skipped": s_skip, "total": s_tests})
+              total_tests += s_tests
+              total_failed += s_failed
+              total_skipped += s_skip
+
+          total_passed = max(total_tests - total_failed - total_skipped, 0)
+
+          if total_tests == 0:
+              header = "## :warning: No test results were produced"
+          elif total_failed == 0:
+              header = f"## :white_check_mark: Tests passed — {total_passed}/{total_tests}"
+          else:
+              header = f"## :x: Tests failed — {total_failed}/{total_tests} failed"
+
+          lines = [marker, header, ""]
+          if suites:
+              lines.append("| Suite | Passed | Failed | Skipped |")
+              lines.append("|-------|--------|--------|---------|")
+              for s in suites:
+                  icon = ":white_check_mark:" if s["failed"] == 0 and s["total"] > 0 else ":warning:" if s["total"] == 0 else ":x:"
+                  lines.append(f"| {icon} `{s['name']}` | {s['passed']}/{s['total']} | {s['failed']} | {s['skipped']} |")
+
+          if failed_cases:
+              lines += ["", "<details open>", "<summary><b>Failed tests</b></summary>", ""]
+              for suite_name, label in failed_cases[:50]:
+                  lines.append(f"- **{suite_name}** — `{label}`")
+              if len(failed_cases) > 50:
+                  lines.append(f"- …and {len(failed_cases) - 50} more")
+              lines += ["", "</details>"]
+
+          lines += ["", f"[View workflow run]({run_url})"]
+
+          with open("comment.md", "w") as f:
+              f.write("\n".join(lines) + "\n")
+          PY
+
+      - name: Upsert sticky PR comment
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const fs = require('fs');
+            const body = fs.readFileSync('comment.md', 'utf8');
+            const marker = '<!-- browseros-agent-tests-summary -->';
+            const { owner, repo } = context.repo;
+            const issue_number = context.payload.pull_request.number;
+
+            const triggerSha = context.payload.pull_request.head.sha;
+            const { data: pr } = await github.rest.pulls.get({ owner, repo, pull_number: issue_number });
+            if (pr.head.sha !== triggerSha) {
+              core.info(`PR head has moved (${pr.head.sha} vs ${triggerSha}) — skipping stale comment.`);
+              return;
+            }
+
+            const comments = await github.paginate(github.rest.issues.listComments, {
+              owner, repo, issue_number, per_page: 100,
+            });
+            const existing = comments.find(c => c.body && c.body.includes(marker));
+            if (existing) {
+              await github.rest.issues.updateComment({ owner, repo, comment_id: existing.id, body });
+            } else {
+              await github.rest.issues.createComment({ owner, repo, issue_number, body });
+            }

packages/browseros-agent/README.md

@@ -75,26 +75,24 @@ packages/
 
 ### Setup
 
-Requires [process-compose](https://github.com/F1bonacc1/process-compose):
-
 ```bash
-brew install process-compose
+# Install dependencies, generate agent GraphQL output,
+# and prepare WXT types for apps/agent
+bun run dev-setup
+
+# Run the apps
+bun run start:server
+bun run start:agent
 ```
 
-```bash
-# Copy environment files for each package
-cp apps/server/.env.example apps/server/.env.development
-cp apps/agent/.env.example apps/agent/.env.development
-cp apps/server/.env.production.example apps/server/.env.production
+`bun run dev-setup` is a preflight command only. It does not manage `.env` files.
 
-# Start the full dev environment
-process-compose up
-```
-
-The `process-compose up` command runs the following in order:
+It runs the following in order:
 1. `bun install` — installs dependencies
-2. `bun --cwd apps/agent codegen` — generates agent code
-3. `bun --cwd apps/server start` and `bun --cwd apps/agent dev` — starts server and agent in parallel
+2. `bun run codegen:agent` — generates agent GraphQL code
+3. `bun run prepare:agent` — generates WXT type output
+
+If you want the broader dev harness after setup, use `bun run dev:watch`, `bun run dev:watch:new`, or `bun run dev:manual`.
 
 ### Environment Variables
 

packages/browseros-agent/package.json

@@ -10,6 +10,7 @@
   ],
   "scripts": {
     "postinstall": "bun run build:agent-sdk",
+    "dev-setup": "bun install && bun run codegen:agent && bun run prepare:agent",
     "dev:watch": "./tools/dev/run.sh watch",
     "dev:watch:new": "./tools/dev/run.sh watch --new",
     "dev:manual": "./tools/dev/run.sh watch --manual",
@@ -26,7 +27,8 @@
     "build:agent:dev": "FORCE_COLOR=1 bun run --filter @browseros/agent --elide-lines=0 build:dev",
     "build:agent": "bun run codegen:agent && bun run --filter @browseros/agent build",
     "build:agent-sdk": "bun run --filter @browseros-ai/agent-sdk build",
-    "codegen:agent": "bun run --filter @browseros/agent codegen",
+    "codegen:agent": "bun --cwd apps/agent graphql-codegen --config codegen.ts",
+    "prepare:agent": "VITE_PUBLIC_BROWSEROS_API=${VITE_PUBLIC_BROWSEROS_API:-http://127.0.0.1:9105} bun --cwd apps/agent wxt prepare",
     "test": "bun run test:tools && bun run test:integration",
     "test:tools": "bun run --filter @browseros/server test:tools",
     "test:integration": "bun run --filter @browseros/server test:integration",