How Google’s Open‑Source Agent Skills Tackle AI Laziness

The article introduces Addy Osmani’s open‑source Agent Skills, a 19‑skill package that embeds Google’s engineering constraints into AI programming tools, providing commands for each development phase and a built‑in “no‑excuse” table to prevent AI from skipping specs, tests, and reviews.

AI Engineering
AI Engineering
AI Engineering
How Google’s Open‑Source Agent Skills Tackle AI Laziness

AI programming assistants often skip specification documents, testing, and security reviews, resulting in code that fails after deployment.

Addy Osmani packaged a Google‑engineered workflow into an open‑source collection called Agent Skills , comprising 19 skills that span the entire development lifecycle.

Define : requirement analysis and PRD writing

Plan : task breakdown

Build : incremental development, API design, front‑end engineering

Verify : test‑driven development, debugging, browser testing

Review : code quality, security hardening, performance optimization

Ship : Git workflow, CI/CD, documentation, release

Each skill embeds Google’s hard constraints—“Spec before code”, “Tests are proof”, and “Faster is safer”—to enforce disciplined development.

7 shortcut commands

/spec

– define what to build /plan – plan the construction /build – incremental build /test – verify functionality /review – conduct code review /code-simplify – simplify code /ship – ship to production

Invoking a command activates the corresponding skill set for that stage of the workflow.

Each skill also contains a “no‑excuse” table that directly counters common AI shortcuts such as “I’ll add tests later” or “The requirement is simple, no doc needed”.

How to use

npx skills add addyosmani/agent-skills

Supported agents include Claude Code (officially recommended), Cursor, Windsurf, GitHub Copilot, and any agent that accepts Markdown.

Repository: https://github.com/addyosmani/agent-skills

AI agentsGitHubdevelopment workflowAddy Osmaniprompt commands
AI Engineering
Written by

AI Engineering

Focused on cutting‑edge product and technology information and practical experience sharing in the AI field (large models, MLOps/LLMOps, AI application development, AI infrastructure).

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.