Game QA — Testing & QA agent for Claude Code
Game QA. Runs the autonomous self-verify loop (build → launch .exe → screenshot → read PNG → verdict).
How to install Game QA
Installs to ~/.claude/agents/melons-melons-agents-game-qa.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/MelonS/MelonS-Agents/HEAD/.claude/agents/game-qa.md -o ~/.claude/agents/melons-melons-agents-game-qa.md Restart Claude Code, or start a new session, for it to be picked up.
What Game QA does
name: game-qa description: Game QA. Runs the autonomous self-verify loop (build → launch .exe → screenshot → read PNG → verdict). Catches regressions across Days. Triggered after every Day's Build Engineer pass. tools: Read, Bash model: sonnet
You are the QA subagent for game-dev-agent.
Role
Empirical truth. Every claim a programmer or designer makes about the build ("the score updates", "the pawn chops", "the menu loads") must be confirmed by a screenshot. No claim trusted witho
Alternatives in Testing & QA
- Witness — Independently witness that an Allium loop's convergence claim is true and was reached honestly 474 ★
- Review Test Structure — Reviews test code for isolation, independence, readability, documentation quality, and proper setup/teardown 432 ★
- Game QA Runner 107 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Kavach Confirm Reporter
KAVACH live-validation reporting agent. Aggregates every confirm_status verdict kavach-poc-executor and kavach
Bead Polisher
Polishes .beads/beads.jsonl in isolated context. Runs the full polish loop (self-containment, dependencies, te
Fe Deck Reader
Decomposes PPT/planning decks (multi-slide, converted to PDF/PNG) into policy & rules / screen mockups / scre
MCP Dogfooder
Drives the shipped MCP debugger tools as a real user would — launch, breakpoints, stepping, locals, object exp
Finalisation Coordinator
Phase 3 INLINE orchestrator: syncs the branch, runs the verify-phase test gate, chatgpt-pr-review, doc-sync sw
Sf Flow Agent
Build, test, and review Record-Triggered, Screen, Scheduled, and Orchestration Flows with approvals and sub-fl
Related Skills
Churn
Run W6 Churn Crisis. Triggered when crisis_severity_score 7 or NRR < 90%. Produces ARR-prioritised interventio
Screenshot Kit
Claude Code skill for automated screenshot & diagram generation. Browser capture + HTML-to-PNG pipeline using
Claudecode Skill HTML Screenshot
HTML → PC + Mobile side-by-side screenshot composer. A Claude Code skill that renders local HTML at desktop an