Who it is forA developer who wants an agent to build features in small, tested slices instead of one giant change.
The problemAgents jump straight to code, skip the spec and produce large changes that are hard to verify.
The outcomeA written spec, a plan of small atomic tasks and code built one slice at a time, each test-driven and committed.
What you need first
- Node.js for `npx skills`, or Claude Code, Codex, Gemini CLI or Antigravity for their native installs
- Git
- SSH keys for the Claude Code marketplace, or the HTTPS URL workaround in the README
Steps
- Run
npx skills add addyosmani/agent-skillsto install all 25 skills into your agent - Run
/specand describe what to build; the spec comes before any code - Run
/planto break it into small, atomic tasks - Run
/buildto implement one slice at a time, or/build autoto implement every task after one plan approval - Run
/testto prove it works before you merge
Prompt for your agent
Using addyosmani/agent-skills, run /spec for <feature>, then /plan, and show me the plan. After I approve, implement it with /build one slice at a time, test-driven, committing each task, and stop on any failing test.
Caveat. /build auto still pauses on failures or risky steps; it removes the stepping between tasks, not the verification.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for addyosmani/agent-skills and "Go from idea to tested feature with /spec, /plan and /build"
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer