DEV Community

Chris Lam
Chris Lam

Posted on

Hermes Core Skills: 25 Executable AI Agent Skills for Debugging, Planning & Token Efficiency

I've been building AI agent tools, and I'm excited to share what I've been working on.

Hermes Core Skills is a collection of 25 executable skills designed for AI agents — Claude Code, Codex, Cursor, and Hermes Agent. Each skill is a standalone SKILL.md file with YAML frontmatter, domain classification, and token cost metadata.

What's included

Systematic Debugging — find and fix bugs faster
Test-Driven Development — write tests before code
Code Review — structured review workflows
Security Hardening — protect your codebase
Token Efficiency — optimize agent context usage
Planning & Decomposition — break down complex tasks
And 19 more...

Quick Start

bash
git clone https://github.com/chrislamlayer1-gif/hermes-core-skills
cp -r hermes-core-skills/skills/* ~/.claude/skills/

That's it. All 25 skills are ready to use.

Why skills?

Skills are the best way to encode agent workflows. Each skill includes:

• Clear triggers (when to use it)
• Step-by-step instructions
• Verification checklists
• Token cost metadata (so agents know upfront what they're loading)

Get involved

The project is MIT licensed and open for contributions. Star it on GitHub, open issues, submit PRs — let's build better AI agents together.

👉 github.com/chrislamlayer1-gif/hermes-core-skills

Top comments (0)