Search

cognitive-os · 认知操作系统 — 13 Skills + 4 Rules + 4 SubAgents，把人脑分层记忆结构外化为 AI 可操作的知识体系，以50年认知科学研究为基础

Record real OpenAI/Anthropic API traffic and emit fakellm.yaml rules automatically.

LLM inference memory trace platform for MRM research

LLM-based code generation and docker-sandboxed execution in a loop in parallel

Lightweight pre-RAG routing layer for LLM applications

AI skill for Claude Code and Gemini CLI — takes a raw product idea through a 5-phase validation framework, producing either a kill rationale or an engineering-ready PRD.

Second-generation project management runtime designed to replace PMC and gstack.

Homebridge platform plugin for Hubspace smart home devices (Home Depot / Afero cloud). Supports lights, ceiling fans, outlets, and switches.

CLI for Pega Infinity DX API V2 (Constellation)

Engineering Playbook — Claude Code skills, metodología SDD+BDD y knowledge base semántica para Next.js

See exactly what your coding agent sends to the model — a local logging reverse-proxy + web dashboard for Claude Code, Codex, DeepSeek-TUI, and Kimi. Just run `ccglass`.

Pi extension package for Feature Factory sub-agents, skills, and workflow orchestration

A lightweight local API proxy providing unified entry for all LLM providers with automatic protocol format conversion

Coding agent CLI with read, bash, edit, write tools and session management

Spec-first clean-room workflow for authorized source analysis without replacement code.

Floating assistant bubble + panel + chat runtime. ESM, no build.

Runtime repair layer for Anthropic tool-calling failures

Session intelligence for Claude Code — active token management via skills and hooks

JSON-RPC server that exposes Claude Agent SDK to any client (codex app-server compatible wire protocol).

Agent skills for git workflows, GitHub automation, security audits, code refactoring, and project tooling