openwifi-mcp

openwifi-mcp

Turns Claude Code into an interactive OpenWiFi triage assistant, providing tools for searching Linux wireless patches, Jira tickets, running LLM-based triage, backporting, and building OpenWiFi.

Category
访问服务器

README

openwifi-mcp

Three MCP servers that turn Claude Code into an interactive OpenWiFi triage assistant. When registered in your project's .mcp.json, Claude gains native tools to search the linux-wireless patch mirror, look up Jira tickets, run the LLM triage pipeline, dry-run + apply upstream backports, and drive OpenWiFi builds — all from within the same conversation you're already having about a bug.

Nothing runs in the cloud. All state lives on your machine or the build host you already control.

What ships

Server Purpose Backing service
openwifi-jira Read-only Jira ticket lookup + similar-ticket search openwifi-ticket-index
openwifi-patches linux-wireless FTS5 search + LLM triage + backport helpers openwifi-kernel-patches
openwifi-build make build/status/tail/fetch/deploy wrappers over SSH Your OpenWiFi build project

All three are stdio-transport MCP servers written in Python with FastMCP.

Tools exposed

openwifi-jira (read-only):

  • jira_list_releases — synced Jira fixVersions with ticket counts
  • jira_search_tickets(q, fix_version?, status?, assignee?, limit?) — full-text search
  • jira_get_ticket(key) — full ticket incl. comments, attachments, changelog, cross-refs
  • jira_get_attachment_text(id) — text content of indexed attachments
  • jira_ticket_similar(anchor, limit?) — similar tickets by key or symptom

openwifi-patches:

  • patches_stats — DB stats + LLM availability
  • patches_search(q, subject_prefix?, author?, since?, until?, limit?)
  • patches_get(message_id) — raw RFC 5322 body
  • patches_expand_query(symptom) — LLM-generated FTS5 query preview
  • patches_triage(symptom | ticket, top_n?) — full pipeline (patches + LLM explanations + similar tickets)
  • patches_dry_run_backport(message_id, tree) — non-mutating apply check
  • patches_apply_backport(message_id, tree) — MUTATING — commits to the tree

openwifi-build — wraps your Claude-OpenWiFi make targets:

  • build_help, build_status, build_tail(tag, lines?), build_fetch(tag) (non-mutating)
  • build_start(branch, target, pr?) — MUTATING — kicks off a long-running build
  • build_deploy(tag, ap, confirm=False) — MUTATING — sysupgrades a real AP

Install

git clone https://github.com/YOU/openwifi-mcp.git
cd openwifi-mcp
./setup.sh

The setup creates a venv, installs the package editable, and writes .mcp.snippet.json you can drop into your project's .mcp.json (or ~/.claude.json for user-scoped access).

Point three env vars at your existing services:

TICKET_INDEX_URL    = http://<host>:20504   # openwifi-ticket-index
PATCH_SEARCH_URL    = http://<host>:20503   # openwifi-kernel-patches
CLAUDE_OPENWIFI_ROOT = /abs/path/to/Claude-OpenWiFi

Restart Claude Code and the tools appear in every session under that project.

Suggested permission allowlist

Add to your project's .claude/settings.json so read-only tools skip the confirmation prompt but mutating tools still require it:

{
  "permissions": {
    "allow": [
      "mcp__openwifi-jira__*",
      "mcp__openwifi-patches__patches_stats",
      "mcp__openwifi-patches__patches_search",
      "mcp__openwifi-patches__patches_get",
      "mcp__openwifi-patches__patches_expand_query",
      "mcp__openwifi-patches__patches_triage",
      "mcp__openwifi-patches__patches_dry_run_backport",
      "mcp__openwifi-build__build_help",
      "mcp__openwifi-build__build_status",
      "mcp__openwifi-build__build_tail",
      "mcp__openwifi-build__build_fetch"
    ]
  }
}

patches_apply_backport, build_start, and build_deploy are deliberately NOT allowlisted — they mutate real state and should always prompt.

Slash command / skill

Pair with a skill like this in .claude/skills/openwifi-triage/SKILL.md so the user can just type /openwifi-triage WIFI-15461 (see the example in Claude-OpenWiFi).

License

MIT. See LICENSE.

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