Build, Evaluate and Ship Agent Skills: Your AI's Superpower

4.6 (10)

Anshumani Ruddra

Product Leader and Super IC at Google

This course is popular

11 people enrolled last week.

Build AI Agent Skills that Actually Work - 3 Hours Hands-on Workshop

Most product builders use AI agents the same way they used search - typing queries and hoping for the best. That's leaving most of the value on the table.

The teams pulling ahead aren't using better prompts. They're teaching their agents. Agent Skills are reusable, testable instruction sets that tell AI agents exactly how to handle specific workflows: your deployment process, your code review standards, your team's edge cases. Write one skill well, and every agent on your team benefits from it, every time, automatically.

The problem: nobody is teaching product builders how to create them. Most documentation is written for a narrow technical audience, and the tooling is fragmented across Claude Code, Codex, and Gemini.

In this hands-on workshop, you'll go from zero to a working skill installed in your own agent: one that reflects your real workflows, not generic templates. You'll also learn how to evaluate and improve skills systematically, so they get better over time instead of drifting.

If you build or are hoping to build products with AI agents: this is the highest-leverage skill you can develop right now.

What you’ll learn

🤖 Learn to build procedural Agent Skills, transforming your professional mindset from solo prompter to manager of elite AI agents

👨‍🏫 Move beyond basic prompting to treat AI as a "PhD-level intern" that requires clear procedural training
🧠 Understand the "token economy" to manage context windows and prevent AI memory loss during long sessions

🛠️ Master the extraction method: turn your manual workflows and successful chats into automated, reusable skills
📂 Use "progressive disclosure" to load complex instructions without bloating the AI's active memory buffer
✅ Add "gotchas" to your skills to document edge cases and failures, making your AI output more robust and reliable

🌐 Navigate platforms like skills.sh to find, evaluate, and install expert-built plugins for marketing and engineering
💪 Customize community templates to fit your specific professional style, bending open-source power to your will

📈 Implement binary scoring and auto-research loops to measure and improve the quality of your AI agents
🚀 Transform your unique expertise into a force multiplier, allowing you to achieve 2x to 10x your current daily output

Workshop agenda

🧠 Why Skills Beat Prompts Every Time
Learn why stuffing instructions into CLAUDE.md breaks at scale, and how the progressive disclosure model behind Agent Skills solves the context window problem for good.
🧭 The Mental Model: Skills vs MCP vs Agents
Cut through the confusion. Understand exactly when to use a skill, an MCP server, a subagent, or an always-on config file — and why getting this wrong wastes weeks of effort.
📘 The Open Standard: Anatomy of a Great SKILL.md
Walk through the structure of a production-ready skill - frontmatter, instructions, gotchas, defaults, and scripts - and see how the same file deploys across Claude Code, Codex, and Gemini.
🛠️ Live Build: Write and Ship Your First Skill
Hands-on exercise. You'll author a real skill for a workflow you own, install it in your agent, and invoke it live. Leave with something working, not just notes.
☕ Break
Short 15 minutes break
📈 Eval-Driven Development: Skills That Improve
Learn how to design binary eval criteria and run the autoresearch loop: an autonomous improvement cycle that takes skills from 50% baseline to 85%+ accuracy overnight.

Learn directly from Anshumani

Anshumani Ruddra

Product Super IC at Google with 22+ years crafting experiences and blockbuster products

Built products at

See all products from Anshumani

Who this workshop is for

Product Managers: who use Claude Code or Codex daily but rely on ad-hoc prompts - and want repeatable, team-shareable AI workflows.
Entrepreneurs & Founders: building AI-native products who need their agents to encode business logic, not just generate code on demand.
Tech Executives & Product Leaders: who want their teams to stop reinventing prompts and start shipping a shared skills library.