god-mode
This skill provides a cross-repo project overview and AI agent coaching, syncing GitHub data and analyzing agents.md. It runs shell commands like god setup, uses CLI authentication (gh auth login/gh auth status), and calls GitHub via gh api (e.g., repos/$repo/contents/$pattern).
god-mode
🔭 Bird's-eye view of your coding projects. 🧠 AI agent coaching based on your actual work.
What is this?
god-mode is an OpenClaw skill that gives you:
- Project Overview - See all your repos at a glance (commits, PRs, issues)
- Agent Coaching - Automatically improve your
agents.mdbased on commit patterns
The Problem
You're juggling multiple projects. You use AI coding assistants but your agents.md instructions are generic or outdated. You lose context when switching between repos.
The Solution
# See everything at once
$ god status
🔭 god-mode
tandem-evaluator
Last: 2h ago • fix: evaluation metrics
PRs: 2 open (1 needs review) • Issues: 5
tada
Last: 1d ago • feat: API endpoints
PRs: 0 • Issues: 3
This week: 23 commits • 3 PRs merged
# Make your AI assistant smarter
$ god agents analyze tandem-evaluator
🧠 Analyzing tandem-evaluator...
⚠️ GAPS FOUND
Testing (not mentioned)
But 31% of your commits touch tests
→ Add: "Write tests for new code"
📝 SUGGESTED ADDITIONS
## Testing
- Unit tests for all new functions
- Run `npm test` before commits
Installation
Prerequisites
gh- GitHub CLI (authenticated)sqlite3- Usually pre-installedjq-brew install jqorapt install jq
Install
# Clone
git clone https://github.com/InfantLab/god-mode-skill
cd god-mode-skill
# Add to PATH
echo 'export PATH="$PATH:'$(pwd)'/scripts"' >> ~/.bashrc
source ~/.bashrc
# Setup
god setup
Or for OpenClaw:
openclaw skills add god-mode
Quick Start
# 1. Add your first project
god projects add github:yourname/yourrepo
# 2. Sync data
god sync
# 3. See the overview
god status
# 4. Analyze your agents.md
god agents analyze yourrepo
Commands
| Command | Description |
|---|---|
god status | Overview of all projects |
god status <project> | Details for one project |
god sync | Fetch latest data from repos |
god sync --force | Full refresh (ignore cache) |
god projects | List configured projects |
god projects add <uri> | Add a project |
god agents analyze <project> | Improve your agents.md |
Configuration
~/.config/god-mode/config.yaml:
projects:
- id: github:myuser/myrepo
name: My Project
priority: high
tags: [work, api]
local: ~/code/myrepo
sync:
initialDays: 90
commitsCacheMinutes: 60
See config.example.yaml for all options.
How Agent Analysis Works
- Reads your
agents.md(or similar file) - Analyzes your commit history (types, patterns, pain points)
- Finds gaps between instructions and actual work
- Suggests improvements based on your patterns
Example insights:
- "You write lots of tests but don't mention testing in agents.md"
- "40% of commits are error-handling fixes - add error handling guidance"
- "Your 'use TypeScript strict' instruction is working - 0 type errors"
Data & Privacy
- All data stored locally in
~/.god-mode/ - No tokens stored - uses your existing
ghauth - No telemetry - nothing phones home
- Open source - audit the code yourself
Roadmap
v0.1.0 (Current)
- Project status overview
- GitHub integration via
ghCLI - Incremental sync with SQLite cache
- Agent instruction analysis
v0.2.0
- Context save/restore
- Activity summaries (
god today,god week) -
god agents generatefor new projects
v0.3.0
- Azure DevOps support
- GitLab support
- Proactive alerts via OpenClaw heartbeat
v1.0.0
- Cross-project intelligence
- Integration ecosystem (Obsidian, etc.)
Contributing
Contributions welcome! Each command is a standalone script - easy to add features.
# Structure
scripts/
├── god # Entry point
├── commands/
│ ├── status.sh # god status
│ ├── sync.sh # god sync
│ ├── projects.sh # god projects
│ └── agents.sh # god agents
└── lib/
├── providers/ # GitHub, Azure, GitLab
└── analysis/ # Commit patterns, agent analysis
See HANDOVER.md for architecture details.
License
MIT - see LICENSE
Credits
Created by InfantLab for the OpenClaw community.
"Know what's happening. Make your AI smarter."