Test — Testing & QA agent for Claude Code
QA role for one type:user-story issue whose E2E trace is already green in CI.
How to install Test
Installs to ~/.claude/agents/cjhowe-us-workflow-test.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/cjhowe-us/workflow/HEAD/agents/test.md -o ~/.claude/agents/cjhowe-us-workflow-test.md Restart Claude Code, or start a new session, for it to be picked up.
What Test does
name: test description: QA role for one type:user-story issue whose E2E trace is already green in CI. Executes the manual test script verbatim via Chrome / Playwright MCP computer-use tools, records PASS/FAIL as an issue comment, opens an iterate spike on FAIL. model: sonnet color: yellow
test
You are the **QA executor** for one glibre `type:user-story` issue whose E2E trace is already green in CI.
MCP loading sequence (REQUIRED)
The `mcp__claude-in-chrome__*` tools are deferred
Alternatives in Testing & QA
- Codex Debugger — Root-cause a failing test, crash, stack trace, or misbehaving feature by delegating the investigation to OpenA 249 ★
- Playwright Test Generator — Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants 246 ★
- Eval Improver — Improver role of the D365FO agent eval loop 136 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
QA Richness
Editorial-density and motion-richness QA gate. Verifies type contrast, accent scarcity, chrome present, not-fl
Manual Tester Agent
Use this agent to test the live application under test the way a human QA would - persona-driven exploration (
Ndv Build
Spec-to-code implementation specialist. Use when a story has defined input/output schemas, acceptance criteria
Worktree Agent QA
Functional verification agent, working in the same git worktree of a User Story already FINALIZED by its imple
Issue Analyzer Fixer
Use this agent when you encounter a bug, error, failing test, or code issue that needs both diagnosis and reme
Issue Subagents Dev
Dev agent for issue-subagents skill. One-shot per invocation — initial spawn implements the spec via TDD again
Related Skills
Verify Kit
Run the kit's mechanical invariant check (bin/verify-kit.sh) and report its three-state output verbatim — neve
Analyze Test Report
Phân tích test report — automation (Playwright/Allure/JUnit/TestNG) hoặc manual (execution_report.md / retest_
Rogue E2e Delivery Skill
End-to-end software house delivery. Use whenever the user asks to build a feature, fix a non-trivial bug, star