Back to Workshops
Workshop Free

Skills.md from Scratch: Build a Skill-Driven Coding Agent

January 16, 2026 Alexey Grigorev
ai-agents llm-engineering agent-systems tooling-architecture

We start from the coding agent from the prerequisite workshop and turn it into a general-purpose coding agent with two reusable behavior layers: skills and slash commands. Skills are loaded by the agent when the user's request matches a skill description. Commands are invoked by the user with a leading slash and rendered into prompts before the agent acts.

Links

External resources for this workshop:

The agent you will build

The final workshop system looks like this:

flowchart LR USER["User"] RUNNER["ToyAIKit runner OpenAI Responses"] LLM["OpenAI model"] CODETOOLS["Coding tools read, write, tree, bash, search"] SKILLTOOL["skill(name) tool"] LOADER["SkillLoader"] SKILLS["skills/*/SKILL.md scripts and templates"] COMMANDS["commands/*.md"] COMMANDTOOL["execute_command(name, args)"] USER -->|plain request| RUNNER USER -->|/command| RUNNER RUNNER --> LLM RUNNER --> CODETOOLS RUNNER --> SKILLTOOL SKILLTOOL --> LOADER LOADER --> SKILLS RUNNER --> COMMANDTOOL COMMANDTOOL --> COMMANDS

The project stays small enough to understand in a notebook, but it mirrors the pieces used by real coding agents. The coding tools let the model read, write, search, and run commands. The skill loader turns SKILL.md files with YAML frontmatter into tool-loadable instructions. The command loader turns markdown files like review.md or test.md into reusable prompt templates.

Appendix

A file list for the workshop materials is in Appendix: workshop files.

Materials

Workshop Code Repository

code

Prerequisite: Building a Coding Agent (Python/Django)

article

Prerequisite: Create a Coding Agent Workshop (GitHub)

code

ToyAIKit Library

code

Related Course: AI Engineering Buildcamp

article

Tutorial pages

  1. Overview and setup
  2. Part 1: Demoing skills and commands
  3. Part 2: Recapping the coding agent
  4. Part 3: Building the base runner
  5. Part 4: Defining the skill format
  6. Part 5: Loading `SKILL.md`
  7. Part 6: The skill tool and prompt injection
  8. Part 7: Implementing commands
  9. Part 8: The fuller prototype
  10. Q&A: skills and commands
  11. Deferred items
  12. Appendix: workshop files

Watch the recording

Full workshop video with timestamps and downloadable materials.

View code on GitHub

Related content

Workshop

Building AI Agents with MCP, PydanticAI and OpenAI

We build a course FAQ assistant from the bottom up. First we expose a plain Python search(query) function to the OpenAI Responses API. Then we turn the same idea into a reusable a…

September 1, 2025

Workshop

Building Safe AI Agents with Guardrails

We start with a DataTalks.Club Data Engineering Zoomcamp FAQ assistant, then add checks that keep the agent on topic, block unsafe responses, and show how to cancel wasted work wh…

January 6, 2026

Article

Inside ChatGPT’s /home/oai/ Folder: Not a Data Leak, but OpenAI’s Hidden Skills System

I unpacked the mysterious /home/oai/ directory that ChatGPT can zip on command and found not leaked OpenAI infrastructure, but a sandbox of document, slide, and spreadsheet “skill…

February 10, 2026