hermes atlas
apr·2026 153·repos hermes·v0.10.0 ★ star this repo

Romanescu11/hermes-skill-factory

A meta-skill plugin for Nous Research's Hermes AI agent that watches your workflows and automatically turns them into reusable skills. Every time you work with Hermes and solve something — setting up a project, debugging code, creating a PR — that workflow disappears at the end of the session. You have to explain it again next time.

★ 339 langPython updated2026-03-18

Skill Factory is a meta-skill for Hermes Agent that automates the creation of reusable skills by monitoring user workflows. It functions by tracking session patterns in the background and proposing structured skill definitions when it detects repeatable sequences. Users can generate both SKILL.md documentation and Python plugin scaffolds directly from these proposals via slash commands or natural language. This tool streamlines the expansion of the Hermes ecosystem by turning manual environment setups and development tasks into permanent, executable assets.

  • Automatically detects and proposes reusable skills from active workflows
  • Generates both SKILL.md documentation and Python plugin scaffolds
  • Provides slash commands to manage, list, and save detected patterns
full readme from github

🏭 Skill Factory

The meta-skill that creates skills. Your AI watches your workflows and turns them into reusable Hermes skills — automatically.

Built for Nous Research's Hermes Agent (v2026.3+).


What It Does

Every time you solve a problem with Hermes, you're performing a workflow worth repeating. Skill Factory watches silently, detects patterns, and at the right moment asks:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🏭 SKILL FACTORY — New Skill Detected
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

I noticed you repeatedly set up a Python environment,
installed dependencies, and ran tests in the same order.

Proposed Skill:   python-env-setup
Category:         software-development
Description:      Reproducible Python project setup workflow

What it captures:
  1. Create venv and activate
  2. Upgrade pip and install dependencies
  3. Run pytest to verify environment

Generate: [A] SKILL.md  [B] plugin.py  [C] Both  [D] Skip
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Say C and it writes both files immediately:

  • ~/.hermes/skills/<category>/<name>/SKILL.md — AI instructions for the workflow
  • ~/.hermes/plugins/<name>.py — A slash command that triggers it directly

Installation

Requirements: Hermes Agent v2026.3+

git clone https://github.com/your-username/hermes-skill-factory
cd hermes-skill-factory
bash install.sh

Or manually:

# Install the meta-skill
mkdir -p ~/.hermes/skills/meta/skill-factory
cp skills/skill-factory/SKILL.md ~/.hermes/skills/meta/skill-factory/

# Install the plugin
cp plugins/skill_factory.py ~/.hermes/plugins/

Then activate:

hermes skills reload
hermes skills enable skill-factory

Usage

Once installed, Skill Factory runs in the background during every session.

Command What it does
/skill-factory propose Analyze the session and propose the top detected skill now
/skill-factory list List all skills generated this session
/skill-factory status Show how many patterns are being tracked
/skill-factory queue Show all detected patterns queued for proposal
/skill-factory save <name> Save the last proposal with a custom name
/skill-factory clear Clear the current session log

Tip: You can also just tell Hermes naturally:

  • "Save this as a skill"
  • "Remember how to do this"
  • "Turn this workflow into a reusable skill"

What Gets Generated

SKILL.md

A complete skill definition following Hermes' native skill format:

---
name: Python Env Setup
category: software-development
description: Reproducible Python project setup
tags: [python, venv, testing]
---

# Python Env Setup

## When to Activate
...

## Workflow
### Phase 1: Environment
1. python -m venv .venv
2. source .venv/bin/activate
...

## Examples
...

plugin.py

A scaffolded Hermes plugin with a slash command:

def register(hermes):
    @hermes.command(name="python-env-setup", ...)
    async def run_skill(ctx, args=""):
        # Step 1: Create venv
        # Step 2: Install deps
        # Step 3: Run tests
        ...

Repo Structure

hermes-skill-factory/
├── skills/
│   └── skill-factory/
│       └── SKILL.md          # The meta-skill (core AI instructions)
├── plugins/
│   └── skill_factory.py      # Plugin: /skill-factory commands
├── templates/
│   ├── SKILL_TEMPLATE.md     # Template for generated skills
│   └── PLUGIN_TEMPLATE.py    # Template for generated plugins
├── examples/
│   └── generated/
│       └── git-pr-workflow/  # Example of a Skill Factory output
│           └── SKILL.md
├── docs/
│   └── how-it-works.md       # Architecture deep-dive
└── install.sh                # One-command installer

How It Works

See docs/how-it-works.md for a full breakdown.

TL;DR:

  1. SKILL.md teaches the Hermes AI how to observe, detect, and propose skills
  2. skill_factory.py provides the /skill-factory commands and file generation
  3. You work normally — Skill Factory watches silently and proposes at the right moment

Examples

The examples/generated/ directory contains real example outputs from Skill Factory:


Contributing

PRs welcome. If you've generated a great skill using Skill Factory, consider adding it to examples/generated/ with a PR.


License

MIT