ianshank/Agents
Agent, Skills, Harness & LangFuse Eval Framwork
Agent, Skills, Harness & LangFuse Eval Framwork
npx skills add ianshank/AgentsAgent, Skills, Harness & LangFuse Eval Framwork
Reusable Anthropic Agent Skills (SKILL.md) packaging real eval methods: clinical-note grounding eval + token-F1 reference scorer. Stdlib-only Python, runs offline.
Multi-Agent Mutual Control, it’s showtime.
Configure custom AI agents for Claude Code without frameworks, runtimes, or additional API costs.
Local-first, open-source AI testing for the web — as a VS Code extension. Hover spawns the coding-agent CLI you already run (Claude Code / OpenAI Codex) to drive your real Chrome via Playwright MCP, then crystallizes clean runs into plain @playwright/test specs that pass CI with zero AI. ✦ optimize pass · 🟠 security · 🔴 pentest
Claude Code skill to support Android app's reverse engineering
TSA for agent skills: scan SKILL.md files before install, deny overreach with readable receipts, and bundle the safe ones.