laguagu/agents-best-practices
.agents-first best practices for AI agent skills — templates, guides, and auditing tools for Claude Code, Codex, Gemini CLI
.agents-first best practices for AI agent skills — templates, guides, and auditing tools for Claude Code, Codex, Gemini CLI
npx skills add laguagu/agents-best-practices.agents-first best practices for AI agent skills — templates, guides, and auditing tools for Claude Code, Codex, Gemini CLI
AI Agent Reference Architecture Patterns
This project demonstrates how to use Claude Code Skill (Form C architecture) to drive a Python anomaly detection pipeline to perform anomaly detection on half-hour granular electricity consumption data from smart meters.
Open-source AI security suite — Shield (Chrome ext) + Sentinel (npm/pip SDK) + Guardian (hosted dashboard) + Claude Code skills. Apache 2.0.
Global ~/.claude/ harness — enforces RPI (Research → Plan → Implement → Closeout) cycles, deterministic project bootstrap, and AI failure tracking via 5 hooks + 4 orchestrator skills + 3 wrapper sub-agents. Cross-platform (Windows/WSL/Linux).
Production-ready personal AI agent runtime in Go with WebUI, multi-agent orchestration, memory, audit, and distributed node support.
Deterministic, risk-governed multi-agent trading engine for US equities. A strategy swarm proposes orders; an independent risk guardian has final say. Paper mode by default, gated live Robinhood MCP execution.