DeepSeek Harness (DeepSeek 官方 agent 浏览器 UI) fnOS 应用 — 本地常驻服务

A long-term memory system built for AI Agents. Agent wakes up already knowing who he is, not querying "who am I?" every session. Every turn calling back accurate memory context. Achieving accurate memory hits while also preventing memory from expanding at scale. No compression, no forgetting.

CLI for SuperDesign Platform - agent skills for Claude Code

Typesafe markdown for AI — compile .skill.md, .agent.md, and .doc.md into Claude Code or Codex outputs

An MCP server to query any Postgres database in natural language.

A Claude Code plugin that provides a conversational, adaptive, and agentic subagent for co-creating rigorous, multi-perspective technical specs—featuring dynamic memory graphs, adaptive workflows, and explicit multi-role analysis. Especially useful for non-technical users and teams adopting the Spec Kit Methodology.

One command to cut token usage by up to 50%+

A Model Context Protocol (MCP) server that retrieves information from Wikipedia to provide context to LLMs.

Multi-model AI code review server using OpenRouter - get diverse perspectives from multiple LLMs in parallel

Metorial MCP Index - An ever growing list of open source MCP servers 📁 🎉

Manage and switch between multiple Claude Code (and other AI coding tool) profiles/accounts locally — isolated config dirs, per-account email, one-command switching.

Vibe coding should have human in the loop! interactive-mcp: Local, cross-platform MCP server for interact with your AI Agent

Use this cross-platform skill in Codex or Claude Code to establish repository-local continuity memory so a future agent can recover objective, status, decisions, validation, risks, and next actions without relying on previous chat history.

🌋 Build AI agents that seamlessly combine LLM reasoning with real-world actions via MCP tools — in just a few lines of TypeScript.

Atlas Cloud skills for Claude Code, Codex & Gemini CLI — generate images/videos and call 300+ AI models from your coding agent.

Enables LLM agents to safely query educational administration databases through read-only MCP tools, enforcing SQL whitelisting, automatic LIMIT caps, parameterized queries, timeouts, token authentication, and audit logging.

Enables AI agents to analyze an SQLite e-commerce database via secure read-only SQL queries, providing tools for table inspection and analytical requests.

Provides a weather lookup tool and greeting resource via MCP, enabling AI agents to fetch current conditions and personalized messages through natural language.

Enables agents to read and resolve layered feedback, propose and make decisions with evidence, and access trial ledgers and judgment views for experiment-driven work.

Provides MCP tools for hybrid knowledge base search, grounded Q&A with citations, agent execution, and ticket/account lookups.

Provides four MCP tools for neural retrieval: personalized user recommendations, item-to-item similarity, cold-start text search, and explainable recommendation scores, enabling LLM agents to query a trained two-tower recommender over Amazon Reviews data.

A security scanner for your LLM agentic workflows

Model Context Protocol server for GraphQL

The most expensive bug in AI-assisted building is shipping the wrong thing, well. A CLAUDE.md for PM + eng teams using Claude Code- 7 principles for problem framing, scope, tradeoffs, and outcome measurement. Built on @karpathy's observations.