SKILLS Refinement Workshop
SKILLS Refinement Workshop
Showcase
Description
Seven processes, one closed loop. SkillForge Refinement Workshop transforms the gstack engineering sprint methodology open-sourced by YC leader Garry Tan into a full lifecycle management system for Claude Code Skills. From six soul-searching questions to verify demand authenticity, to architecture planning, SKILL.md forging, quality review, eval testing, release packaging, and finally review and iteration—each stage has dedicated AI roles, with confidence scores and anti-spam mechanisms. It also includes two additional processes: emergency diagnosis and security audit. No 'good ideas,' only evidence-based judgments. Every skill is tempered through seven processes.
Related Skills
View all
AFP .skill Forging Engine v4
> Turn a one-sentence idea into an AFP skill people actually want to install—full-process forging. 🌍 Niche reconnaissance — scan similar skills before starting (GitHub, ClawHub, skill market), benchmark horizontally and vertically to find real differentiation. No niche statement, no start. 🧬 Three-layer nested architecture — governance, cognitive, and execution layers separate skeleton from flesh; four-quadrant admission judgment, actively discourage forbidden zones, don't overspec simple tasks. 🔬 Dual-core adversarial audit — A-core builds + B-core dismantles, 9-dimension scoring matrix (including release readiness), ratchet lock ensures scores only increase, single-variable control prevents disorder from over-editing. 📦 Full asset packaging for listing — naming formula + subtitle + icon + golden description structure + 3 intro images + 4-channel promotion copy + 15-item listing checklist + direct callApi creation. After opening, just tell it what you want to do (e.g., 'Turn my writing method into a skill,' 'Optimize my existing XX skill,' 'Upgrade the old AFP to 3.1'). The engine will guide you from niche reconnaissance all the way to packaging for listing. Each phase has deliverables, quality gates, and stop points. > v3.1 meta-architecture refactoring: replaced the old five-stage process with a three-layer nested skeleton, added niche reconnaissance and horizontal/vertical competitor benchmarking, 9-dimension audit matrix and ratchet mechanism, cross-conversation state persistence, and fully retained full-asset packaging capability.
Skill Quality Audit (PDCA-QMS)
Evaluate Skills using a quality management system based on management science, not an off-the-cuff checklist. Based on Six Sigma CTQ/FMEA/DPMO × TQM × Deming PDSA cycle, through the Plan-Do-Check-Act four phases, it provides: a quality scorecard with Sigma level, a defect priority table ranked by RPN, root cause analysis down to the instruction design layer, and the full Skill text after Poka-Yoke correction. The fixer and reviewer are forcibly separated, and scores only increase, never decrease. v2.0 additions: ① Quantitative contract – closed enumeration of structural unit counting (only phase level), K-value gradient anti-scoring lock, Sigma table lookup with log axis interpolation, and full zero-padding to prevent division by zero, ensuring scores are reproducible and undistorted; ② Quality level × disposal path dual-axis gate, eliminating the contradiction of 'unqualified but recommended for release' labels; ③ Diagnostic mode – say 'only a report' or 'don't modify yet', then run through P-D-C and stop at Check, without modifying your script; ④ Output contract – pass/fail points fold, evidence limited to 50 characters, phase word count limits, values must include calculation process. When to use: evaluate whether a Skill is good, diagnose why its output is unstable, systematically optimize an existing Skill, benchmark multiple Skills, pre-release quality inspection of Skill drafts, or only want a diagnostic report without modifying the draft. Trigger words: detect skill quality, skill quality check, quality audit, rate a skill, PDCA optimize skill, skill health check, evaluate this skill, skill defect analysis, diagnose only without repair, pre-release skill quality check, skill benchmarking.
AI Lesson Forge
Welcome to [AI Lesson Forge]! This is not a cold program that simply hands you a bunch of tool links, but a red-blue adversarial training ground full of strategic tension. Here, every teaching idea of yours will undergo an extremely rigorous 'stress test under real conditions': 🔴 Red Team (Education Experts + AI Experts): Your super think tank. Based on your teaching pain points, they tailor cutting-edge AI intervention plans (e.g., auto-generating multimodal courseware, building professional scenario-based training agents, etc.). 🔵 Blue Team (Dynamic Adaptive Opponent): The 'soulful contrarian' of our forge. The system dynamically matches the opponent based on your education level—for K-12, they become students who fear boredom and parents who fear trouble; for higher education, they become academic ethics supervisors who delve into professional boundaries and clinical risks. Your plan only passes if it survives their scrutiny! ⚪ Referee (That's you): Sitting in the role of the teaching research lead, you synthesize the debate outcomes and make the final call on the implementation roadmap. Just drop in a chapter outline and the pain points that trouble you, and the forge will output an intuitive 'Pain Point Heatmap' and a structured 'Adversarial Evaluation Report' with one click. Whether you want to add fun to your classroom or conduct complex scenario simulations with high professional barriers, [AI Lesson Forge] can help you keep quality and avoid pitfalls. Let us ignite the vitality of educational reform through red-blue adversarial play, and uphold the original aspiration of education with digital empowerment! 🍵✨
Information
- Version
- v2
- Last updated
- Runtime credits
- Usage-based
- Models
- Auto
