GitHub
v1.0.6
管理代码库、用户、发布版本,并自动化 GitHub 工作流
使用场景/代码审查与 GitHub
拉取 PR、读 diff、查 Issue、管理仓库。适合 Code Review、发布说明、CI 问题排查。
共匹配 44,428 个资源 · 第 64 / 926 页
GitHub MCP 让 AI 进入你的协作流程:拉取 PR diff 做代码审查、检索 Issue 历史定位已知问题、读取 CI 日志分析构建失败、汇总提交生成 changelog。官方 github/github-mcp-server 通过 GitHub API 或个人访问令牌(PAT)鉴权,Gitee / GitLab 也有对应实现。
Code Review 实践:把团队的评审规范写成 Skill(如「检查竞态、边界、错误处理,按严重度输出」),配合 GitHub MCP 让 AI 先过一遍 PR,人工只看 AI 标出的高风险行——审查耗时通常明显下降,且低級错误漏检更少。
权限建议:使用细粒度 PAT(只授予目标仓库 contents / pull-requests 读写),组织仓库务必开只读起步;让 AI 发布评论或合并前先要求确认,防止误操作污染协作记录。
v1.0.1
gbXML MCP: parse buildings, spaces and surfaces, validate structure and areas.
v1.0.1
EAGLE MCP: parse schematics and boards, validate parts, nets and sheets.
v1.0.1
Duck Creek MCP: parse XML manifests, validate structure and duplicate names.
v1.0.1
CIF MCP: parse crystal data blocks, tags and loops with column checks.
v1.0.1
CellML MCP: parse models, validate components, variables and units.
v1.0.1
Calypso MCP: parse trade export CSV, validate columns, amounts and dates.
v1.0.1
Buck2 MCP: parse BUCK files, validate targets, loads and names.
v1.0.1
Autotools MCP: parse configure.ac and Makefile.am, validate macros and sources.
v1.0.1
ASN.1 MCP: parse modules, list type assignments and validate structure.
v1.0.1
ARINC 429 MCP: decode label, SDI, data, SSM and odd parity from 32-bit words.
v1.0.1
Adabas Natural MCP: parse programs, validate data blocks and variables.
v1.0.3
Ab Initio MCP: parse graphs, validate components, ports and edges.
v0.1.10
Drive FL Studio from an AI agent with one Python tool: write notes, mix, render and listen.
v1.0.1
AAF MCP: parse AAF-XML content storage, mobs and tagged values.
v1.1.0
Check Solana wallets and transactions for risky approvals and scam tokens; find SOL to take back.
v0.2.2
Person and company profiles, company search and logos, work emails and email checks
v3.3.0
Turn music ideas, articles and tracklists into playable mixes, cited stories and narrated clips.
v0.9.0
Annotate screenshots for user guides. Saved as plain JPEG/PNG whose annotations stay editable.
v0.4.2
Local video editor: cut silence, edit by transcript, add captions, export to Final Cut or Premiere.
v4.10.0
Hire humans for digital microtasks, research, QA, data evaluation, and physical-world tasks.
v0.9.2
Author custom Power BI visuals on your live Desktop model: preview, self-review, commit to PBIP.
v0.31.0
Connect Runner to Claude or ChatGPT, then ask about your business in a chat.
v0.4.0
Read-only packet capture analysis with tshark: Diameter, GTP, SIP, RTP, SIGTRAN, DNS, TCP and TLS.
v0.1.13
Labs64 NetLicensing 的 MCP 服务器,用 AI 管理软件许可证生命周期。
v0.1.1
Tech stack, WHOIS, DNS, SPF/DMARC, SSL expiry and AI crawler rules per domain. Runs on Apify.
v0.1.1
Bulk HTTP status checker: 404s, 5xx and redirect chains for thousands of URLs. Runs on Apify.
v0.1.1
Technical SEO audit crawler: broken links, 404s, redirects, titles, canonicals. Runs on Apify.
v0.1.1
Bulk PageSpeed Insights and Core Web Vitals (LCP, INP, CLS) for many URLs. Runs on Apify.
v0.1.1
Track brand mentions and citations in ChatGPT, Perplexity, Claude and Grok answers. Runs on Apify.
v0.2.1
Selenium tests from user stories (Python or Java) and self-healing locators with an audit trail.
v1.1.0
Open-source CRM: Claude, ChatGPT or any MCP client works with records, inbox and company knowledge.
v1.0.7
创建、投放并分析广告活动,洞察竞争对手——Google、Meta、TikTok、LinkedIn 等。
v1.1.1
Discover Agentel Missions, Skills and public Work evidence; continue through authorized SDK/API.
v1.4.0
Query India's open geo data, and author collect maps: register a link, moderate, publish.
v0.2.0-beta.3
Work tracker for teams: create, triage, and update issues, projects, and milestones as yourself.
v0.1.10
Claude Code or Codex read and edit your open Figma file. Needs the free FigmaMCP Mac app.
v0.9.30
面向 AI 智能体的真实邮件与短信——发送邮件、接收验证码、管理真实收件箱。
v1.43.1
Econometrics and causal inference (DiD, IV, RD, synthetic control, DML) with Stata/R parity grades
v2026.11.5
Search US trademarks, screen proposed names for conflicts, and look up USPTO registration records.
v0.1.25
Work a running OrcaSlicer with AI: slice, tune, and understand what each setting costs
v1.5.0
Self-hosted agent memory in an Obsidian vault: hybrid search, link graph, tasks, safe writes, OAuth
v0.2.1
User stories to verified Playwright tests, and self-healing locators with an audit trail.
v2.3.0
Persistent memory MCP server for Claude Code, Codex and OpenCode, stored locally in SQLite
v0.57.5
面向 Obsidian 仓库的独立 MCP 服务器——混合搜索、笔记与文件、记忆、任务、OAuth 2.1。
v1.15.47
Your AI mind directs coding agents; work reviewed against your memories: Claude Code, Codex, Cursor
v1.4.0
x402 machine-payable utility edge: 19 curated MCP tools + ~450-SKU REST catalog, priced in USDC
v1.0.2
Let AI agents watch videos and answer questions about them through any OpenAI-compatible endpoint.
v1.2.2
Search Angela Kamer's coastal house plans by beds, sq ft, lot width, pilings and elevator.
本地 stdio 部署用个人访问令牌(推荐细粒度 PAT)最简单;多用户托管场景才需要 GitHub App / OAuth。环境变量名见资源详情页。
会。它是第一道过滤而非替代:语义类缺陷检出不错,但供应链、越权逻辑仍需专业审计与人工评审兜底。
取决于模型提供方与令牌权限。企业内网可选本地模型 + 只读令牌;令牌务必设置过期与最小 scope,并定期轮换。