Skip to content

snapsynapse/skill-provenance / skills/bootstrap

bootstrap

Unverified

Add version tracking to an existing unversioned skill bundle. Use when you have a skill project with SKILL.md but no MANIFEST.yaml, no version headers, and no changelog. Creates the initial manifest, adds version metadata to files, and generates the first changelog entry. Triggers on: "bootstrap this skill", "add versioning", "start tracking versions", "create a manifest", "this skill has no versioning".

Skill Provenance: Bootstrap

Add provenance tracking to an existing unversioned skill bundle. This is a one-time operation that creates the initial manifest, version headers, and changelog.

When to use

Run /skill-provenance:bootstrap when you have a skill directory that contains a SKILL.md but lacks MANIFEST.yaml. This is the entry point for adopting Skill Provenance on an existing project.

If the bundle already has a MANIFEST.yaml, use /skill-provenance:open instead.

Protocol

  1. Inventory all files. Read the directory structure yourself — do not ask the user to list files. Identify:

    • SKILL.md (required — if missing, this isn't a skill bundle)
    • Evals files (e.g., evals.json)
    • Scripts (e.g., generate.js, validate.sh)
    • Output artifacts (.docx, .pdf)
    • Source material provided by the user
    • Any other files present
  2. Propose version numbers. Ask the user what version to assign. If there's a handoff note, git history, or other context suggesting prior versions, propose a number based on that history. Default to 1.0.0 for the bundle and 1 for individual files if no history exists.

  3. Add internal version headers to files that can safely carry YAML frontmatter. Use the metadata block pattern for SKILL.md:

    ---
    name: skill-name
    description: What the skill does.
    metadata:
      skill_bundle: skill-name
      file_role: skill
      version: 1
      version_date: 2026-04-11
      previous_version: null
      change_summary: Initial versioned release.
    ---
    

    For files that cannot safely carry frontmatter (.json, .sh, binaries), the manifest tracks their version — no modification needed.

  4. Determine frontmatter mode. Maintain the canonical source in frontmatter_mode: metadata when richer provenance is useful. For Codex, Gemini CLI, or another strict loader, derive a minimal-frontmatter install copy rather than weakening the canonical source in place. If the user explicitly wants one strict-only bundle, minimal is valid.

  5. Generate MANIFEST.yaml. Include:

    • bundle, bundle_version, bundle_date, description
    • compatibility section based on user's target platforms
    • Exactly one top-level files: list using unquoted normalized relative paths and a sha256: hash or explicit hash: null for every entry
    • Leave deployments empty unless the user mentions existing installs.
  6. Generate CHANGELOG.md. Create a single entry summarizing known history. If the user has context about prior changes, incorporate it. Otherwise:

    # Changelog
    
    ## 1.0.0 — 2026-04-11
    - Initial versioned release. All files inventoried and hashed.
    - SKILL.md: [brief description of current state]
    - evals.json: [N evals covering X]
    
  7. Deliver the versioned bundle. Provide all new and modified files. Summarize what was created and any decisions for the user to confirm.

File roles

Assign these roles in the manifest based on file type:

RoleFiles
skillSKILL.md
evalsevals.json, evals-*.json
script.js, .py, .sh scripts
output.docx, .pdf rendered artifacts
handoffhandoff notes
sourceuser-provided material (tracked, not versioned)
referencedocumentation in references/
assettemplates, images, fonts in assets/

Conventions referenced

  • Version identity lives inside files when their format allows it, and always in the manifest.
  • MANIFEST.yaml is not self-listed. It tracks other files only.
  • Per-file versions are integers. Bundle version is semver.
  • Hashes are SHA-256 of file contents.
  • Filenames never contain version numbers. Version lives in metadata and the manifest, not in SKILL_v5.md.

For the full spec: https://skillprovenance.dev

rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 467kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 196kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 196kUpdated Claude CodeCodexVerifiedInstall