maggy
Turn Claude Code into a self-reviewing, multi-model engineering system
All Skills in This Repository (59)
Pure Emerald Level IndicatorsExternal Model Delegation Pattern
Routes user prompts to tiers via UserPromptSubmit hook and delegation scripts.
cpg-analysis
Identify control-flow, data-flow, and taint-driven security issues via Joern and CodeQL.
maggy
Convert issue tracker backlogs into prioritized execution sessions with test enforcement.
Visual Validation โ Autonomous Screenshot Verification
Compares Peekaboo screenshots with AI to detect visual regressions.
icpg
Track code changes against formal contracts using a Reason Graph.
code-graph
Navigate AST-based code graphs for symbol lookup and dependency analysis.
cross-agent-delegation
Route engineering changes to Codex, Kimi, or Claude by complexity and risk.
Autonomous Testing Agent
Automate test coverage discovery, generation, execution, and repair for Python and TypeScript projects.
polyphony
Provision per-task Docker workspaces with dedicated git branches for parallel agent execution.
Model Routing System
Route user prompts to AI model tiers with cascading classifier fallbacks.
mnemos
Maintain a typed MnemoGraph memory store to restore context after compaction.
Build in Public โ Best Practices
Create LinkedIn and X posts from engineering milestones with anonymization rules.
Frequently Asked Questions
FAQPage SchemaHow to install Maggy?โผ
Run `npx skills add alinaqi/maggy --all -g -y` in your terminal to install all skills in this suite globally.
What does Maggy do for Claude Code?โผ
It adds TDD enforcement, quality gates, code review hooks, and cross-session memory so Claude Code ships tested, reviewed code instead of unverified output.
How does Maggy reduce AI coding costs?โผ
A local classifier scores each task and routes it to the cheapest capable model, sending simple work to free or low-cost models and reserving Claude for architecture and security.
Does Maggy work with Codex, Kimi, and Gemini CLI?โผ
Yes. It orchestrates delegation and review across Claude, Codex, Kimi, Gemini, DeepSeek, and local models through shared memory checkpoints and routing hooks.
Can I run multiple AI agents in parallel safely?โผ
Yes. The Polyphony system runs each agent in its own Docker container with an isolated git branch, so parallel work never causes file conflicts.
Related Repositories in Software Engineering
View All in Software Engineeringโopenclaw
Run a personal AI assistant across your devices and chat apps
superpowers
Gives coding agents a disciplined workflow from idea to merged code
react
AI agent skills for building, testing, and porting React core