Public skills are reviewed before promotion. Draft and experimental workflows stay in the incubator until they have enough evaluation proof and a clear maintenance path.

Quick start

List the public catalog without installing anything:

npx skills@latest add stark-ai-de/agent-skills --list

Install one skill into the agent detected for the current project:

npx skills@latest add stark-ai-de/agent-skills --skill architecture-compass

Update installed skills:

npx skills@latest update

npx skills@latest add stark-ai-de/agent-skills --skill codex-memory-curator codex-spec-interviewer animated-readme-logo architecture-compass codegraph-ast-grep drawio-diagrams -g -a codex -y

npx skills@latest add stark-ai-de/agent-skills --skill cursor-memory-curator cursor-spec-interviewer animated-readme-logo architecture-compass codegraph-ast-grep drawio-diagrams -g -a cursor -y

npx skills@latest add stark-ai-de/agent-skills --skill claude-memory-curator claude-spec-interviewer animated-readme-logo architecture-compass codegraph-ast-grep drawio-diagrams -g -a claude-code -y

Each bundle contains the skills from its matching *-operations category plus every public skill outside the operations categories. The bundles install globally; remove -g for a project-local install. Avoid --skill '*' when targeting one runtime because it also selects skills built specifically for other runtimes.

The -a option selects the host where a skill is installed; it does not change that skill's target contract. An intentionally cross-host install preserves target-specific evidence and output while adapting collaboration controls to the selected host.

Choose a skill

When you need to…	Use	Scope
Turn a fuzzy coding request into an implementation-ready spec	`codex-spec-interviewer`, `cursor-spec-interviewer`, or `claude-spec-interviewer`	Runtime-specific
Audit or clean persistent agent context	`codex-memory-curator`, `cursor-memory-curator`, or `claude-memory-curator`	Runtime-specific
Set up ADR guardrails or align a refactor with accepted decisions	`architecture-compass`	Cross-runtime
Combine semantic impact analysis with syntax-exact structural search	`codegraph-ast-grep`	Cross-runtime
Create, edit, validate, or export an editable diagram	`drawio-diagrams`	Engineering workflow
Create or transform an accessible README logo pipeline	`animated-readme-logo`	Engineering workflow

See the complete public catalog for exact trigger descriptions and skill-evals/ for maintainer proof.

How the catalog works

Path	Purpose	Published?
`skills/`	Reviewed skills discoverable through `npx skills`	Yes
`incubator/skills/`	Candidate, experimental, or not-yet-public skills	No
`skill-evals/`	Activation cases, comparisons, and promotion evidence	Maintainer proof only
`site/`	Astro source for the generated catalog	GitHub Pages

Promotion is a folder move from incubator/skills/<category>/<skill>/ to skills/<category>/<skill>/, backed by demonstrated value, activation tests, a reusable use case, manageable maintenance, and reviewable evaluation proof. See the spec policy for public-contract updates.

Development

pnpm install
npm run validate

Command	Purpose
`npm run list`	List promoted public skills
`npm run list:incubator`	List incubator skills
`npm run validate`	Validate skills, ADRs, scripts, and repository contracts
`npm run smoke:install`	Test clean-copy discovery and exact host destinations
`npx skills@latest add ./skills --list`	Check local public skill discovery
`pnpm --filter ./site build`	Build the generated catalog
`pnpm format:check`	Check formatting
`pnpm lint`	Lint repository scripts
`node scripts/validate-release.mjs`	Validate release readiness

Scaffold a skill only when its destination is clear:

npm run scaffold repo-maintenance/my-new-skill
npm run scaffold:incubator engineering-workflows/my-candidate-skill

Documentation

Topic	Reference
Naming, safety, and review rules	Contributing
Validation and CI	Validation
Publishing and release flow	Publishing
Specs and durable decisions	Spec policy · ADR index
Domain language and boundaries	Glossary · Out of scope
Logo animation and fallback behavior	Motion specification

Compatibility and safety

This repository follows the open Agent Skills specification. Individual compatibility varies by skill and host; read a skill's SKILL.md before installing it. Runtime-specific skills stay in their runtime category, while reusable workflows live under engineering-workflows.

Skills are executable context. Helper scripts are read-only unless their documentation explicitly says otherwise. Never add secrets, tokens, customer data, private repository paths, or internal hostnames to public skills, references, assets, tests, or examples.

License

Apache-2.0 for the skills and repository material published here.

Agent Skills

README