# addyosmani/agent-skills

Production-grade engineering skills for AI coding agents.

Repository: https://github.com/addyosmani/agent-skills
Canonical: https://ross.abutalabs.com/products/agent-skills
Homepage: https://skills.addy.ie
Language: JavaScript
License: MIT
License Family: permissive
Topics: agent-skills, antigravity, claude-code, cursor, skills, codex
Last push: 2026-08-21T18:15:04+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 98, longevity 14
- inputs: {"age_days": 199, "days_push": 12, "days_rel": 19, "gap_med": 15.5, "n_releases_24m": 9}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 90043, forks 9630 (observed 2026-08-28T04:12:23.219972+00:00)

## What it is
A curated pack of 24 production-grade engineering skills (structured Markdown workflows) for AI coding agents like Claude Code, Cursor, Codex, and Copilot. It provides eight lifecycle slash commands (/spec, /plan, /build, /test, /review, /webperf, /code-simplify, /ship) that encode senior-engineer workflows, quality gates, and best practices across the software development lifecycle.

## Use cases
- make my AI coding agent follow senior-engineer best practices
- write a PRD spec before coding with an AI agent
- get consistent code reviews from Claude Code or Cursor
- run test-driven incremental implementation with an AI agent
- audit web performance and Core Web Vitals with my agent
- harden code against OWASP vulnerabilities via AI agent workflows
- simplify and clean up messy code with an AI assistant
- automate the full define-to-ship development lifecycle with slash commands

## When to choose
- you use Claude Code, Cursor, Codex, Copilot, Cline, or any agent that accepts Markdown instructions and want structured engineering workflows
- you want consistent quality gates for spec, build, test, review, and ship phases
- you prefer installing individual skills or the whole pack via the open skills CLI

## When to avoid
- you need executable tooling or a runtime library rather than Markdown workflow instructions
- your agent cannot read Markdown rules or skills files
- you want opinion-free, minimal prompting without prescribed processes

## Facets
- artifact type: plugin
- maturity: active
- function: developer-tools, code-review, testing, security, workflow-automation, prompt-engineering
- domain: developer-tools, web-development
- platform: cli, cross-platform
- tags: ai-coding-agents, claude-code, cursor, codex, skills, sdlc, slash-commands, markdown-skills, spec-driven-development, ai-agents, automation

## Member repositories
- addyosmani/agent-skills (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:23.219972+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:13:51.618289+00:00, confidence not recorded.
  - readme: https://github.com/addyosmani/agent-skills (fetched 2026-08-28T04:12:23.219972+00:00, sha 062c2cf4251e)
  - homepage: https://skills.addy.ie (fetched 2026-08-28T17:36:54.555143+00:00, sha c65fda3c1450)
  - site_page: https://skills.addy.ie/docs/getting-started/ (fetched 2026-08-28T17:36:54.563771+00:00, sha 214e1fd32328)
  - site_page: https://skills.addy.ie/skills/security-and-hardening/ (fetched 2026-08-28T17:36:54.565931+00:00, sha 654dcb5031c9)
- Data as of 2026-08-30T08:39:29.467469+00:00.