hayabusa-mcp

hayabusa-mcp

Enables Windows Event Log (EVTX) analysis by wrapping Hayabusa, exposing scan and rule retrieval tools.

Category
访问服务器

README

hayabusa-mcp

An MCP (Model Context Protocol) server that wraps Hayabusa for EVTX (Windows Event Log) analysis, exposing scan_evtx and get_hayabusa_rules tools, plus read-only detection:// resources for browsing the bundled Sigma rule set (see Resources below).

Note: this README covers licensing and the Claude Desktop extension build step. Full setup/usage documentation is still in progress — see HANDOFF.md for detailed background in the meantime.

Setup

  1. ./download_hayabusa.sh — downloads and checksum-verifies the Hayabusa binary into ./hayabusa/
  2. pip install -r requirements.txt (or use a .venv)
  3. Connect via .mcp.json (Claude Code) or as a Claude Desktop extension (see below)

Resources

Alongside the two tools, the server exposes four read-only detection:// MCP resources for browsing the bundled Sigma/Hayabusa rule set and looking up MITRE ATT&CK technique coverage:

  • detection://rules — a compact JSON index of the bundled rules (resource_id, id_source, title, level only — not full rule bodies). Capped at 500 entries per read; truncated/total_rules in the response indicate when more exist.
  • detection://rules/{rule_identifier} — the complete YAML text of one rule. rule_identifier is the rule's own YAML id: (a UUID); a small number of rules missing a valid id get a stable fallback-<hash> identifier instead — check id_source in the index to tell them apart.
  • detection://rules/by-technique/{technique_id} — rules tagged with a given MITRE ATT&CK technique ID (case-insensitive). A bare parent ID like T1003 matches its own tag plus any sub-technique tagged beneath it (labeled inherited_subtechnique in the response); a specific sub-technique query like T1003.001 matches only that exact tag, never its parent or sibling sub-techniques.
  • detection://attack/techniques/{technique_id} — local coverage facts for a technique ID (match count, covered/not_covered), computed solely from the bundled rule set. This server does not bundle an ATT&CK technique-name/tactic/description dataset, so no such fields are returned — the response only contains what can be derived from local rule data.

Claude Desktop extension

The extension's manifest.json points PYTHONPATH at a vendored ./lib/ directory rather than any external virtualenv. lib/ is not tracked in git (it contains compiled, platform-specific wheels) — build it locally with:

./package_extension.sh

This regenerates ./lib/ from requirements.txt and produces dist/hayabusa-mcp.zip. You'll still need ./hayabusa/ populated separately via download_hayabusa.sh before the extension can run.

License

This project is licensed under the MIT License.

Hayabusa itself is a separate project licensed under AGPL-3.0. It is downloaded at setup time by download_hayabusa.sh and invoked as an external subprocess — it is not vendored or linked into this repository.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选