cheng80

Auto-review pipeline — reads the full CEO, design, eng, and DX review skills from disk and runs them sequentially with auto-decisions using 6 decision principles. Surfaces taste decisions (close approaches, borderline scope, codex disagreements) at a final approval gate. One command, fully reviewed plan out. Use when asked to "auto review", "autoplan", "run all reviews", "review this plan automatically", or "make the decisions for me". Proactively suggest when the user has a plan file and wants to run the full review gauntlet without answering 15-30 intermediate questions. (gstack) Voice triggers (speech-to-text aliases): "auto plan", "automatic review".

Cross-model benchmark for gstack skills. Runs the same prompt through Claude, GPT (via Codex CLI), and Gemini side-by-side — compares latency, tokens, cost, and optionally quality via LLM judge. Answers "which model is actually best for this skill?" with data instead of vibes. Separate from /benchmark, which measures web page performance. Use when: "benchmark models", "compare models", "which model is best for X", "cross-model comparison", "model shootout". (gstack) Voice triggers (speech-to-text aliases): "compare models", "model shootout", "which model is best".

Performance regression detection using the browse daemon. Establishes baselines for page load times, Core Web Vitals, and resource sizes. Compares before/after on every PR. Tracks performance trends over time. Use when: "performance", "benchmark", "page speed", "lighthouse", "web vitals", "bundle size", "load time". (gstack) Voice triggers (speech-to-text aliases): "speed test", "check performance".

Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with elements, verify page state, diff before/after actions, take annotated screenshots, check responsive layouts, test forms and uploads, handle dialogs, and assert element states. ~100ms per command. Use when you need to test a feature, verify a deployment, dogfood a user flow, or file a bug with evidence. Use when asked to "open in browser", "test the site", "take a screenshot", or "dogfood this". (gstack)

Post-deploy canary monitoring. Watches the live app for console errors, performance regressions, and page failures using the browse daemon. Takes periodic screenshots, compares against pre-deploy baselines, and alerts on anomalies. Use when: "monitor deploy", "canary", "post-deploy check", "watch production", "verify deploy". (gstack)

Claude Code CLI wrapper for non-Claude hosts - three modes. Review: independent diff review via claude -p. Challenge: adversarial failure-mode review. Consult: ask Claude about the repo with read-only file tools. Use when asked for "claude review", "claude challenge", "ask claude", "second opinion from claude", or "outside voice". (gstack)

Restore working context saved earlier by /context-save. Loads the most recent saved state (across all branches by default) so you can pick up where you left off — even across Conductor workspace handoffs. Use when asked to "resume", "restore context", "where was I", or "pick up where I left off". Pair with /context-save. Formerly /checkpoint resume — renamed because Claude Code treats /checkpoint as a native rewind alias in current environments. (gstack)

Save working context. Captures git state, decisions made, and remaining work so any future session can pick up without losing a beat. Use when asked to "save progress", "save state", "context save", or "save my work". Pair with /context-restore to resume later. Formerly /checkpoint — renamed because Claude Code treats /checkpoint as a native rewind alias in current environments, which was shadowing this skill. (gstack)

Run the Rummi Poker NAS web build/deploy workflow and verify the deployed site matches the local build. Use when the user says "나스웹빌드 해줘", "나스 웹 빌드", "NAS 웹 배포", "rummipoker NAS deploy", or asks to run tools/deploy_rummipoker_web.sh and then check remote hashes.

Safely commit current repository changes with a caveman-commit style Conventional Commit message and push the current branch. Use when the user says "커밋 푸시", "커밋/푸시", "commit push", "caveman commit push", or asks to commit and push with a terse commit message.

Deprecated alias for caveman-commit-push. Use caveman-commit-push for safe commit and push with a terse Conventional Commit message.

Run and evolve the Rummi Poker leveling/ML pipeline: fresh Dart simulator data collection, chunked contest_policy_v1 runs, economy audit, feature table generation, supervised model smoke reports, and safe tracked report updates. Use when the user says "레벨링 ML", "fresh 데이터", "밸런스 ML", "학습 데이터 쌓기", "시뮬 데이터 수집", or asks to continue the internal leveling data pipeline.

Record real Rummi Poker Flutter web runtime interactions with Playwright video capture, especially fixture-based UI/effect validation such as special tile settlement, boss constraint overlays, market animation, or battle feedback. Use when the user asks to make or verify an actual implementation video, webm, mp4, GIF-like proof, or runtime effect recording.

Create clean technical infographic documents for this project, including README visual summaries, internal RAG structure posters, architecture maps, and beginner-friendly system journey diagrams. Use when asked to make an infographic, visual README, RAG 내부 구조, or example-image-style technical poster.

Project-specific regression checkpoint for the accessible tourism chatbot. Use when asked to recheck remaining tourism work, rerun tourism chat evals, verify noisy/v4/context regressions, update progress docs, or decide whether a tourism code change is safe without manual visual review.

Run the Jewel Match NAS web build/deploy workflow and verify the deployed site matches the local build. Use when the user says "나스웹빌드 해줘", "나스 웹 빌드", "NAS 웹 배포", "jewelmatch NAS deploy", or asks to run tools/deploy_match_web.sh and then check remote hashes.