agentleFS
svy04svy04/ballast/skills/skill-forge/SKILL.md

skill-forge

Skill73 starsChanged 3 months agoMITView on GitHub

Promote a procedure that was worked out end-to-end and verified into a reusable skill file, so the next similar task starts from the solved path. Use when a hard-won procedure succeeds with evidence, when the same multi-step task appears a second time, or when the user says "make this repeatable".

49 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use sigma, agent-forge and 3 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • sanyuan0704Skill

    sigma

    Personalized 1-on-1 AI tutor using Bloom's 2-Sigma mastery learning. Guides users through any topic with Socratic questioning, adaptive pacing, and rich visual output (HTML dashboards, Excalidraw concept maps, generated images). Use when user wants to learn something, study a topic, understand a concept, requests tutoring, says 'teach me', 'I want to learn', 'explain X to me step by step', 'help me understand', or invokes /sigma. Triggers on: learn, study, teach, tutor, understand, master, explain step by step.

    Also in 3 projects that use this

  • xyreinsurance119Skill

    agent-forge

    Создание, улучшение и аудит скиллов и агентов OpenClaw. Три режима: создание скилла (11 шагов), создание агента (9 шагов, с памятью и автоулучшением), улучшение существующего (5 шагов). Triggers: 'создай скилл', 'новый скилл', 'создай агента', 'новый агент', 'улучши скилл', 'skill creator', 'agent creator', 'скиллмейкер'.

    Also in 3 projects that use this

  • localSummerSkill

    skill-review

    回顾当前对话,整理 skill 使用情况、代码质量反馈和迭代建议,生成待确认的更新清单

    Also in 3 projects that use this

  • NVIDIASkill

    deep-research

    Queue deep, multi-step research and analysis tasks to the DeepAgents worker. Supports execution depth presets, request-specific rubrics, SubAgent delegation, and RubricMiddleware cross-validation.

    Also in 3 projects that use this

  • GoogleSkill

    code-review

    Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.

    Also in 3 projects that use this

The file

Files
SKILL.md

Skill forge

Some tasks are solved once and then re-solved forever: the deploy that took an afternoon of trial and error, the migration that only worked in one specific order, the flaky test that needed three environment tweaks. The solving was real work; letting it evaporate means paying for it again. When a procedure has been driven from start to finish and the success was checked, not declared, forge it into .claude/skills/<name>/SKILL.md — the next similar task starts where this one ended.

When to forge

Both conditions, not either:

  1. It will recur. The task shape is likely to come back — same project or same class of problem. One-off procedures stay in the session log.
  2. Success was verified. A check passed: the build is green, the endpoint responds, the output was inspected. "It seemed to work" forges nothing — a skill built on an unverified run teaches the next session a guess.

What goes in the file

The skill file records the procedure as it actually ran, not as it should have run in theory:

  • When to use — the task shape that triggers it, specific enough that the wrong task won't match.
  • Procedure — the steps in the order that worked, including the traps hit along the way (step 3 fails silently if the cache wasn't cleared first is the most valuable line in the file).
  • Verified — the date the procedure last succeeded.
  • Evidence — what confirmed success: the test that passed, the file that existed, the output that was checked. This line is what separates a skill from a hopeful note.
  • Source — the memory/knowledge/ entry the procedure rests on, when there is one. Include this line only in that case; it is what makes the third freshness trigger below reachable.

Frontmatter follows the usual shape: name matching the directory, description saying what it does and when to use it.

Relation to pin

pin promotes a one-line rule — a correction or preference that should arrive with matching messages. skill-forge promotes an entire procedure — multiple steps, ordering, traps. When a lesson fits in a sentence, pin it; when it only fits in a walkthrough, forge it. A forged skill often yields one pin as a by-product: the single sharpest trap, delivered even when the skill isn't opened.

Freshness

A procedure is a claim that these steps still work. Three triggers force a refresh: the verification date is more than 90 days old, the procedure fails when followed, or the knowledge entry this procedure rests on changed. Any of them, re-run it, fix what changed, and update the date and evidence — a skill that silently stopped working is worse than no skill, because it carries the authority of a past success.


Part of ballast — foundations first, every solved path kept, done means checks pass.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.