agentleFS
AsiaOstrichAsiaOstrich/EngramGraph/.agents/skills/skill-builder/SKILL.md

skill-builder

Skill4 starsChanged 29 days agoMITView on GitHub

[UDS] Turn a repeated manual process into a properly scoped Skill, with the right amount of process along the way. Use when: the same multi-step sequence has been done manually three or more times, formalising an ad-hoc Skill, deciding where a Skill belongs. Not for: recording historical facts or project state — that belongs in memory, not a Skill; one-off tasks that will not recur. Keywords: skill, skill builder, process knowledge, repeated process, automation, 技能建立, 流程知識, 自動化.

74 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use literature-research, commit-and-push and 7 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.
  • nickjwellsSkill

    literature-research

    Search academic literature and perform deep research reviews. Use when user asks to search PubMed, find academic papers, or do literature reviews.

    Also in 3 projects that use this

  • nopandererSkill

    commit-and-push

    Automatically analyze git changes, generate Conventional Commits compliant commit messages, commit changes, and push to remote in a single automated workflow. Use when the user invokes /commit-and-push or explicitly requests to automatically commit and push changes following Conventional Commits format (feat:, fix:, docs:, etc.). This skill provides a fully automated git workflow without manual intervention.

    Also in 3 projects that use this

  • MicrosoftSkill

    skill-reviewer

    Review skill PRs with structured severity-rated feedback covering token budgets, routing conflicts, and repo conventions. WHEN: \"review skill\", \"review skill PR\", \"review skill changes\", \"check skill quality\", \"skill PR feedback\".

    Also in 3 projects that use this

  • SkillPilotCHSkill

    level-up

    Eine Automatisierung pro Woche finden, scopen, shippen. Output landet als proposed-Eintrag in cadence.md mit konkreten Implementations-Schritten.

    Also in 3 projects that use this

  • MicrosoftSkill

    excalidraw-diagram

    Create Excalidraw diagram JSON files that make visual arguments. Use when the user wants to visualize workflows, architectures, or concepts.

    Also in 4 projects that use this

  • SentrySkill

    onboard

    Onboard a service to sentry-options. Use when integrating a new service, creating schemas, or setting up the multi-repo sentry-options workflow.

    Also in 5 projects that use this

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 2 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 65 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Before creating SKILL.md, decide where it belongs:

    line 64
  2. Distinctive:

    Not yet recorded → add row, fill current count

    line 83
  3. Very specific:

    Run /sdd to create XSPEC. Return to Step 5 after XSPEC is Approved.

    line 101
  4. Very specific:

    Add ## MODIFIED Requirements or ## ADDED Requirements at the end.

    line 106
  5. Very specific:

    Verify frontmatter: name, scope, description, allowed-tools

    line 111
  6. Distinctive:

    After completing all steps, verify:

    line 133

“Common” is measured: kept by at least 6 in 10 of the 65 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Process-to-Skill Assistant

Language: English | 繁體中文 | 简体中文

Guides you from "I keep doing this manually" to a properly built Skill, with the right amount of process along the way.

When to Use

  • You've done the same multi-step sequence ≥ 3 times manually
  • A teammate asks "how do we do X again?" for the 3rd time
  • You built a Skill ad-hoc and want to formalize it

Core Principle

A Skill records process knowledge. Memory records historical facts. When you notice yourself executing the same steps repeatedly, that's a Skill candidate.

Decision Tree

New Skill needed?
├── Modifying existing Skill?
│     → Delta path: append ## MODIFIED / ## ADDED to existing SKILL.md
│                   update version field → done
│
├── Answer these 4 questions (any "yes" → Complex):
│     1. More than 7 steps?
│     2. Branching logic between steps (if/else)?
│     3. Requires knowledge from 3+ separate standards/decisions?
│     4. Output directly affects sub-project source code?
│
├── All "no" → Simple path
│     → Fill Skill Brief (templates/SKILL-BRIEF-TEMPLATE.md)
│     → Create SKILL.md directly (no XSPEC needed)
│
└── Any "yes" → Complex path
      → Create XSPEC first → run /sdd
      → Return here after XSPEC Approved

Deprecating a Skill?
  → Add to SKILL.md frontmatter:
      status: deprecated
      deprecated_at: YYYY-MM-DD
      deprecated_reason: "..."
      superseded_by: "/new-skill"   (if applicable)
  → Mark archived in SKILL-CANDIDATES.md

Placement Decision

Before creating SKILL.md, decide where it belongs:

ConditionPlacement
Steps reference project-specific paths (e.g. TECH-RADAR.md, DEC-*.md)Project: {project}/.claude/skills/
Steps are universal (no project-specific paths)UDS: skills/{name}/ + zh-TW locale

Workflow

Step 1 — Describe the Process

Capture the repeated sequence:

  • What steps, in what order?
  • How many times done manually so far?
  • What tools or files does it touch?

Step 2 — Update SKILL-CANDIDATES.md

Open your project's SKILL-CANDIDATES.md (copy from templates/SKILL-CANDIDATES.md if first time):

  • Not yet recorded → add row, fill current count
  • Already recorded → increment count
  • Count reached 3 → mark trigger ✅, proceed

Step 3 — Choose Path (Simple / Complex / Delta)

Answer the 4 judgment questions. Determine: Simple, Complex, or Delta.

Step 4a — Simple: Fill Skill Brief

Use templates/SKILL-BRIEF-TEMPLATE.md:

  • Trigger context (when would you use this?)
  • Core steps (3–7, ordered)
  • Acceptance Criteria (2–3 items)
  • Out of scope (explicit boundaries)

Step 4b — Complex: Create XSPEC

Run /sdd to create XSPEC. Return to Step 5 after XSPEC is Approved.

Step 4c — Delta: Identify Change Scope

Identify which sections of the existing SKILL.md change. Add ## MODIFIED Requirements or ## ADDED Requirements at the end.

Step 5 — Create / Update SKILL.md

Generate SKILL.md from the Brief or XSPEC:

  • Verify frontmatter: name, scope, description, allowed-tools
  • For UDS skills: create zh-TW locale version too
  • For project skills: place in {project}/.claude/skills/{name}/SKILL.md

Step 6 — Update SKILL-CANDIDATES.md

Mark the candidate row: trigger ✅, Skill column filled.

Step 7 — Commit

feat(skills): Add /{skill-name} skill. 新增 /{skill-name} Skill。

{English description, 1-2 lines}

{Chinese description, 1-2 lines}

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Output Checklist

After completing all steps, verify:

  • SKILL-CANDIDATES.md updated (trigger ✅, Skill name filled)
  • SKILL.md created with complete frontmatter (name/scope/description/allowed-tools)
  • Simple path: Skill Brief referenced or preserved
  • Complex path: XSPEC ID noted in SKILL.md header comment
  • UDS skill: zh-TW locale file created
  • Deprecated: status: deprecated in frontmatter
  • git commit completed

References

  • Skill Brief template: templates/SKILL-BRIEF-TEMPLATE.md
  • Candidate tracking: templates/SKILL-CANDIDATES.md (copy to your project)
  • ADR standards: core/adr-standards.md

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.