Pipeman — Testing & QA agent for Claude Code
Use this agent to push code to the remote repository after QA1's first gate passes, and to push follow-up fixes during the LiveQA live-test loop.
How to install Pipeman
Installs to ~/.claude/agents/chrishobbsrocks-fully-completely-pipeman.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/chrishobbsrocks/fully-completely/HEAD/.claude/agents/pipeman.md -o ~/.claude/agents/chrishobbsrocks-fully-completely-pipeman.md Restart Claude Code, or start a new session, for it to be picked up.
What Pipeman does
name: pipeman description: Use this agent to push code to the remote repository after QA1's first gate passes, and to push follow-up fixes during the LiveQA live-test loop. Use only after QA1 sign-off, never before, except for a trivial fix per CLAUDE.md's fast lane. model: sonnet color: green
You are Pipeman, who manages git and pipelines for this operation.
CRITICAL BOUNDARIES:
- You do NOT write application code (that's the dev teams' job)
- You do NOT review code for correctness (t
Alternatives in Testing & QA
- Test Fix Loop Agent — Автоматическое тестирование REST API в цикле с исправлением ошибок 571 ★
- Witness — Independently witness that an Allium loop's convergence claim is true and was reached honestly 474 ★
- Writ Reviewer — Reviews an implementation diff in two passes (spec-compliance first, then code quality) 187 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Liveqa
Use this agent to test the live, deployed product in a real browser after Pipeman has pushed a sprint's code.
AI Team Qa.Agent
AI QA engineer agent (Ivy). Use when: testing features, running E2E tests, playtesting, filing bug reports, wr
Research Reba
Guardian/QA - Validates everything, guards safety rails, nothing merges without her. Use for code review, vali
04 QA Engineer
Test planning, execution, bug reports, launch sign-off, quality gates. Nothing ships without your sign-off.
Pixel Critic
Visual QA for pixel art. Use PROACTIVELY before telling a user a sprite is finished, and whenever they ask "is
Raider Test Writer
Write Langertha::Raider tests with Test2::V0 — agent loop, MCP tool calling, CLI/Hall/ACP behavior. Tests neve
Related Skills
Sprint Reship
Pipeman: push a fix during the LiveQA live-test loop
Factory Release Manual
Human-invoked, one-shot release of a factory campaign branch into (never main) — refuse if locked, re-confirm
Check Standards
Check oryxflow project code against the house standards for readable, reliable code - naming (task / column /