Claude Desktop 설치 가이드
로컬 MCP 설정을 지원하는 데스크톱 클라이언트.
MCP Servers
Index a dataset once, then retrieve column profiles, filtered rows, server-side aggregations, and cross-dataset joins — so a million-row file costs thousands of tokens instead of millions.
MCP server for interacting with Apache Superset, enabling AI agents to connect to and control a Superset instance programmatically.
Shared MCP server for the Upvote.club Public API.
A self-modifying, self-verifying AI agent with persistent state. Reads its own source code. Plans changes. Tests them in a sandbox before applying.
One command. TypeScript Python. Example tools, the Inspector, and tests — all wired up. Or point it at an OpenAPI spec and get
- 🧾 — every MCP tools/call request is now recorded in a dedicated, self-hosted table (wp_wsp_mcp_audit_log): tool name, timestamp (UTC), acting user (or unauthenticated), request IP, and the outcome...
MCP Fundamentals 🧱 Meet users in the AI apps they already use Your users are increasingly getting accustomed to using natural language to communicate with AI apps like ChatGPT, Claude Desktop, and...
Add vision to text-only models in Opencode (, ) — , no manual file saving.
MCPMate is a progressive MCP management center for organizing servers, clients, profiles, capabilities, and runtime visibility in one local workspace.
Outcome-Based Persistent Memory MCP Server
The open-source trust layer for bringing Claude Code (and other autonomous AI agents) to work — with your security team's approval.
Four small MCP tools that let apfel (Apple's on-device LLM) reach the web and read local files: url-fetch (read a page), ddg-search (web search), search-and-fetch (search+read in one call), fs...
Agent Skills
Flexible and powerful framework for managing multiple AI agents and handling complex conversations
"一器而工聚焉者，车为多。" ——《考工记》 两千年前，造一辆马车是最复杂的系统工程：**承载乘者，辕定方向，辐传动力，軎辖为约束——每个构件各司其职，合而为一，车方能行。 今天，构建一个 AI Agent 亦是如此：对话循环为**，工具系统为**，权限管线为**，而将这一切承载于其上、使智能体真正运转的运行时框架—— Agent Harness——正是那个**。
*Tuned for (1M-token MSA context, native multimodal input) and (Agents Window, canvases, Design Mode, /worktree, /best-of-n, Await, MCP Apps). Written to stay useful across model changes.*
humanizer-zh 是一个兼容 Codex、Claude Code 和 OpenClaw 的中文去 AI 味技能。它用于重写、润色或审阅中文博客、评论、产品分析、newsletter、书稿章节等长文本，让文字更像中文母语者写出来的，而不是翻译腔、模型拼接稿或营销通稿。
A meta-skill that designs domain-specific agent teams, defines specialized agents, and generates the skills they use.
Official Compound Engineering plugin for Claude Code, Codex, Cursor, and more
Welcome to New Expensify: a complete re-imagination of financial collaboration, centered around chat. Help us build the next generation of Expensify by sharing feedback and contributing to the code.
的 AI Agent 工作台，一定程度上解决Openclaw的长期记忆及安全问题： 1. 对话与证据落盘为本地 JSON； 2. 长期记忆由 memory_module_v2 做结构化蒸馏与混合检索实现11倍tokens压缩； 3. 接入自己微调的Qwen3.5 4b模型配合Langchain的before_agent middleware用于合规内容检测； 4.
基于 Claude Code / Codex 的 SRC 漏洞挖掘 Agent 技能体系 —— 系统级提示词 + 14 个专项安全测试 Skill 知识库
67 Specialized Skills for Full-Stack Developers. Transform Claude Code into your expert pair programmer.
A catalogue of — CRUD tables, detail & master-detail pages, kanban, calendar, wizard, charts, billing, RBAC, i18n, and more — that an AI agent into a real back-office.
Most popular OSS agent harnesses (Cline, Roo Code, Continue, OpenCode, Aider) leave on the table because their prompt-caching code is subtly wrong, off-by-default, or just missing for some providers.