KasperChenGH/token-diet
Claude Code skill: cut agentic-workflow token usage ~85% by optimizing architecture before artifacts — agent census, session merging, compute eviction, scripts-compute-LLM-judges
Claude Code skill: cut agentic-workflow token usage ~85% by optimizing architecture before artifacts — agent census, session merging, compute eviction, scripts-compute-LLM-judges
npx add-skill KasperChenGH/token-dietClaude Code skill: cut agentic-workflow token usage ~85% by optimizing architecture before artifacts — agent census, session merging, compute eviction, scripts-compute-LLM-judges
🌐 Build a Model Context Protocol (MCP) server for seamless integration with LLM applications, enhancing automation and efficiency in workflows.
Enable AI agents to generate structured, brand-consistent DOCX, XLSX, PPTX, and PDF documents with a clear workflow and built-in QA.
AI Agent用のSKILLやSub Agent集
A cognitive simulacrum of Jeremy McEntire — an AI agent that thinks like Jeremy on novel situations. Two-phase classifier dispatch over Anthropic + (optional) OpenAI fine-tune. Includes a deployable Fly.io service, a CLI / Claude Code skill, and a primer for building one for a different subject.
Claude Code skills for medical research — literature search, reporting guidelines, statistical analysis, publication figures. Built by a physician-researcher, tested on real publications. MIT licensed.
AI Native SecOps platform powered by LangGraph multi-agent orchestration. Automated alert triage, pluggable skills, and human-in-the-loop command execution — all in one React WebUI.