ИИ-агенты
ghidra-mcp
Ghidra MCP server with 200+ tools that let AI assistants drive real reverse-engineering work
★ 4,9 тыс. звёздApache-2.0
Смотреть руководство →excel-mcp-server — проект с открытым исходным кодом из категории «ИИ-агенты» от haris-musa (haris-musa/excel-mcp-server на GitHub). У него 4 216 звёзд на GitHub и 470 форков, он распространяется по лицензии MIT и последний раз обновлялся на GitHub 8 октября 2026.
Excel MCP Server: let AI agents read, write, format, and chart real .xlsx workbooks — no Excel required
Choose the folder(s) the server may access — access is confined to --allow-dir paths.
Add the uvx command to Claude Code, Codex, or another MCP client using the documented one-liners.
Restart so the Excel tools load.
Have the agent read a sheet, compute values, and write results or a chart back into the file.
Read/write ranges, formatting, conditional formatting, data validation, tables, images, and sparklines.
Formula results are calculated by a built-in engine (~260 functions), not left as unevaluated strings.
Create Excel charts including 2016 types, and build PivotTables from your data.
Folder confinement, formula safety checks that reject network/external-reference functions, atomic saves, and an optional read-only mode.
Run locally per-client over stdio, or as a streamable-HTTP server with optional token authentication.
Reads VBA in macro-enabled files; writing VBA is opt-in (--allow-vba-write) and macros are never executed.
Clone the project and enter its directory (most users run it via uvx, next step).
git clone https://github.com/haris-musa/excel-mcp-server.git
cd excel-mcp-serverRegister the server, scoped to your workbooks folder.
claude mcp add excel --scope user -- uvx excel-mcp-server stdio --allow-dir /path/to/workbooksStart the server over stdio yourself, or over HTTP at http://127.0.0.1:8017/mcp.
uvx excel-mcp-server stdio --allow-dir /path/to/workbooksDeveloper haris-musa maintains the project (excelmcpserver.com).
Yes, MIT-licensed. Your AI client (Claude, Codex, etc.) may have its own costs.
Neither — no Excel installation is required, and the server itself needs no key; HTTP mode can optionally use your own auth token.
This server gives the agent purpose-built, safety-checked tools (confinement, formula checks, atomic saves, a real calculation engine) instead of arbitrary code execution against your files.
Run `claude mcp add excel --scope user -- uvx excel-mcp-server stdio --allow-dir /path/to/workbooks`, restart Claude Code, and ask it to inspect a workbook.
Ghidra MCP server with 200+ tools that let AI assistants drive real reverse-engineering work
Chrome DevTools for coding agents: an MCP server that lets agents debug and inspect real Chrome
GitHub's official MCP server — let AI tools work with repos, issues and pull requests
Native MCP plugin for x64dbg — let an AI drive the debugger: breakpoints, stepping, memory, registers
Production-grade MCP server for Godot — let AI build scenes, edit scripts and wire your game live
Official MathWorks MCP server: drive MATLAB from AI coding agents.
Другие проекты из категории «ИИ-агенты» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если excel-mcp-server вам не подходит.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Соседние проекты из категории «ИИ-агенты» в каталоге, ближайшие по числу звёзд к excel-mcp-server — продолжайте изучать полку по обе стороны от него.
AdalFlow: The library to build & auto-optimize LLM applications.
Open-source, vision-first browser agent
ArcBox: a fast macOS-native container and VM runtime — a drop-in Docker engine with agent sandboxes
Basic Memory: local-first MCP memory so you and your AI tools build a shared knowledge graph in Markdown
Monid — OpenRouter for agent tools: one base URL and one key give an agent 2,000+ tools across 72+ providers
Langroid: a principled Python framework for building LLM multi-agent applications (from CMU/UW researchers)
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.