Requires durable citations when authoring or researching Agent Skills—maintain references/sources.md per skill, link external research, and record provenance in PRs. Use when creating skills, updating SKILL.md, doing web research for skills, or auditing missing sources.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Rule: Knowledge borrowed from docs, papers, other repos, or chat research must be saved as links in the skill tree—not only in conversation.
skills/{name}/SKILL.mdMaintain skills/{name}/references/sources.md:
# Sources
| Topic | URL | Used for |
|-------|-----|----------|
| Agent Skills spec | https://agentskills.io/ | Frontmatter rules |
| … | … | … |
## Changelog
- 2026-05-29: initial citations
Template: citation-template.md.
references/sources.md (can start sparse).references/ for long material.references/sources.md.”pnpm run validate warns if sources.md is missing.| Cite | Example |
|---|---|
| Official specs | agentskills.io, Cursor/Claude/Codex docs |
| Papers / research | SkillOpt, SkillsBench arXiv |
| Reference implementations | product MCPs, Codex plugin-eval |
| Blog/guides | Only if procedure is non-obvious; prefer primary docs |
Do not cite ephemeral chat URLs or private tokens.
Guild-wide references: docs/STANDARDS.mdx, DESIGN_FAQ.mdx.
| Task | Skill |
|---|---|
| Scaffold skill | skill-authoring-lifecycle |
| Evaluate / improve | skill-eval-improve |
| Spec audit | skill-authoring-lifecycle |
npx skills add arenukvern/skill_steward --skill skill-source-citations
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Design, implement, and integrate generalized validation harnesses across a producer-consumer boundary after a local harness contract exists. Use when refactoring custom validation CLIs/MCPs for large polyrepos, extending Steward across sibling repos, or deploying a local tool to a consumer project for dogfooding and testing; use mcp-harness-repo-maintainer first for initial steward.yaml adoption and cold-start proof.
日本語の概要は準備中です。原文の説明を表示しています。
Maintains repo-local action contracts and harness repositories where product CLI and MCP adapters stay thin over core libraries. Use when adopting or improving steward.yaml actions, capability-level adoption evidence, cold-start proof loops, probes, benchmarks, CLI/MCP/core parity, adapter refactors, packages/core boundaries, or sibling harness layout; use repo-quality-system-lifecycle first for general app/library/tool stewardship baselines.
日本語の概要は準備中です。原文の説明を表示しています。
Run a Mixture of Experts (MoE) audit on any topic, plan, codebase, evidence archive, or process. Dynamically spawns specialized subagents with different critical lenses to cross-reference findings and detect flaws, overlap, retention issues, or drift. Use when designing architectures, analyzing complex code, verifying multi-step plans, classifying evidence artifacts, or looking for duplicated intent in a repo.
日本語の概要は準備中です。原文の説明を表示しています。
Plan and document handoffs, parent lane contracts, and parallel batch contracts between specialized AI agents (foreman, workers, reviewers). Use for multi-agent workflows, subagents, original goal preservation, native gates, claim ceilings, terminal states, baton passes, or guild-style agent coordination.
日本語の概要は準備中です。原文の説明を表示しています。
Designs public or private Agent Skill and plugin marketplaces for Cursor, Claude Code, Codex, Zed, Open Plugin, and npx skills—manifest layout, install matrix, and Skill Steward vs product boundaries. Use when setting up a marketplace, distributing skills/plugins to a team, private registry, .cursor-plugin, .codex-plugin, .plugin, .claude-plugin, or skills.sh publishing.
日本語の概要は準備中です。原文の説明を表示しています。
Chooses ecosystem-native release and changelog tooling (Changesets, Melos, release-plz) plus binary distribution (GitHub Release tarballs, install.sh) when the product is an executable. Use for release CI, install.sh, versioning, CHANGELOGs, shipping MCP/CLI without clone, or meta repos that only ship skills via npx skills—not domain app tutorials.
日本語の概要は準備中です。原文の説明を表示しています。