IndyDevDan · Video summary

RAW Agentic Coding: ZERO to Agent SKILL

Build agent skills from scratch by focusing on core principles: planning, structure, and progressive disclosure of information via a cookbook.

IndyDevDan · December 8, 2025 · 50-min episode · 6-min read

Key takeaways

🚀Before building, define the desired output and purpose of your agent skill on paper.01:45
🎯Begin agent development by planning and defining the concrete output and end-state, rather than immediately prompting.01:45
🔧Organize your agent code into a standard structure: `skill.md`, `tools`, `prompts`, and a `cookbook` directory.03:42
💡Structure agent skills with a clear hierarchy: a central `skill.md` file, a `tools` directory for scripts, a `prompts` directory, and a `cookbook` for detailed instructions.03:42
📝Implement a cookbook within your skills to progressively disclose information and guide agent behavior for specific tasks.05:02
📈The cookbook pattern enables progressive disclosure of information to agents, providing specific instructions and examples for different use cases.05:02

Chapters

0:00🎯 Introduction and Core Philosophy
1:13🎯 Planning and Structuring the Skill
3:19🎯 Building the Basic File Structure
4:26🎯 Prompt Design and Tool Integration
9:10🎯 Implementing the Fork Terminal Skill
20:42🎯 Testing and Iterating the Skill
24:02🎯 Enhancing the Setup with Context and Help Commands
44:01🎯 Scaling and Forking Contexts
47:41🎯 Managing Multiple Agents and Sessions
49:04🎯 Final Tests and Review
49:34🎯 Conclusion and Future Plans

The digest

High-Level Overview

This content presents a detailed walkthrough of building an agentic coding skill from the ground up, specifically a tool that enables an AI agent to fork new terminal sessions and execute CLI commands or spin up new coding agents. Designed for software engineers and AI developers, it emphasizes fundamental engineering principles—planning first, then coding—while showcasing how to construct reusable, composable skills for AI agents without relying on existing orchestration frameworks. The skill enhances agent capabilities by fostering modularity, context management, and parallel execution, key for scalable AI-assisted development workflows.

Keep reading in The Digest

The full 6-minute article, 12 more sections, and every new episode from IndyDevDan and the creators you follow.

Structured Deep Dive
1. The Importance of Skills and Starting with Engineering Fundamentals (0:00–3:00)
2. Designing the Folder and File Structure (3:00–6:00)
3. Defining the Skill’s Purpose and Core Use Cases (6:00–9:00)
4. Implementing the Skill: Prompt Writing and Coding (9:00–22:00)
5. Incorporating Documentation via the Cookbook for Different Tooling (22:00–30:00)
6. Enhancing Agent Behavior: Pre-Execution Help and Smart Command Use (30:00–39:00)
7. Scaling Up: Forking Multiple Agent Contexts and Summarizing Conversations (39:00–48:00)
8. Finalizing the Skill and Reflections on Agentic Skill Building (48:00–50:18)
Key Framework: The RAW Agentic Skill Development Process
Practical Implications & Takeaways
Action-Oriented Summary

Read the full digest — free

More from IndyDevDan

GPT-5.5 VERIFIED Opus 4.7: A Pi Coding Agent That REVIEWS Like YOU

In this video, IndyDevDan introduces a powerful two-agent system combining Opus 4.7 and GPT 5.5, emphasizing the importance of a verifier agent that enhances trust and efficiency in agentic coding by validating claims made by the primary agent. He highlights the need for multi-agent orchestration to maximize the potential of AI models and improve engineering workflows.

May 4, 2026

MAXIMIZE Your Claude Code Subscription (Without Getting BANNED)

In this video, IndyDevDan explains how to maximize your Claude subscription while avoiding bans by adhering to the crucial rule of individual use: if anyone else benefits from your subscription, switch to an API key. He outlines three tiers of usage—safe, controversial, and bannable—to help engineers navigate the complexities of subscription terms.

April 27, 2026