Skip to main content
Glama

20 Best GitHub MCP Servers, Compared (October 2026)

Updated 20 of 1,950 GitHub servers in the Glama registryRebuilt weekly

The short answer

Most readers should start with GitHub MCP Server (github/github-mcp-server). It is GitHub's official MCP Server, documents a hosted HTTP endpoint for repository, issue, and pull request context, had 33,668 stars and 654 stars gained per 30 days, and its default branch had a commit 3 days ago with 171 commits in the last 12 weeks. If whole-file reads waste context in a large repository, github-codemunch-mcp (jgravelle/jcodemunch-mcp) is the better answer: it exposes 7 tools, its tool descriptions are graded A, it had 10,808 PyPI downloads in a typical week, and its default branch had 1,149 commits in the last 12 weeks. For sprint and roadmap work in GitHub Projects V2, mcp-github-project-manager (kunwarVivek/mcp-github-project-manager) is the better answer.

Whichever you choose, give it the narrowest access that still works (a read-only credential, a replica, a scratch account), and widen it only once you have watched what your agent actually asks for.

Glama operates the MCP registry these numbers are measured from, and sells MCP hosting and a gateway. No position on this page is paid for. How the registry is built.

Quick picks

  • 1GitHub MCP Server : A developer with a remote-capable MCP host who needs GitHub repository, issue, and pull request context: it documents a hosted HTTP endpoint.
  • 2github-codemunch-mcp : Exploring a large GitHub repository where whole-file reads waste context: it indexes source once, then serves targeted symbols through seven tools.
  • 3mcp-github-project-manager : Planning sprints and roadmaps for teams that live in GitHub Projects V2: it exposes milestone, sprint, and project tools alongside issue and pull request management.
  • 4Octocode MCP : Researching code across a local checkout and GitHub repositories: it exposes 13 tools for local ripgrep/LSP search plus GitHub code, repo, PR, and package lookup.
  • 5idea-reality-mcp : Before starting a new project: it exposes one idea_check tool that scans GitHub, Hacker News, npm, PyPI, and Stack Overflow for existing competitors.

Which one, for your situation

Your situationWhat to use
You need repository, issue, and pull request context over HTTPUse GitHub MCP Server (github/github-mcp-server): it documents a hosted HTTP endpoint and its default branch had 171 commits in the last 12 weeks.
You need token-efficient symbol search in a large repositoryUse github-codemunch-mcp (jgravelle/jcodemunch-mcp): it exposes 7 tools, its tool descriptions are graded A, and its default branch had 1,149 commits in the last 12 weeks.
Your team plans sprints and roadmaps in GitHub Projects V2Use mcp-github-project-manager (kunwarVivek/mcp-github-project-manager): it exposes 84 tools for projects, milestones, tasks, and sprints, and its default branch had 56 commits in the last 12 weeks.
You need duplicate code checks before committing a snippetUse jscpd (kucherenko/jscpd): it exposes 4 tools, its tool descriptions are graded A, and its default branch had a commit 0 days ago.
You want static analysis for Python, TypeScript, or Go codeUse mcp-skylos (duriantaco/skylos): it exposes 16 tools for static analysis, its tool descriptions are graded A, and it had 15,530 PyPI downloads in a typical week.
You need review before destructive git, SQL, or shell actionsUse Node9-Proxy (node9-ai/node9-proxy): it checks tool calls and holds destructive git, SQL, and shell actions for review, and its default branch had a commit 0 days ago.

Top MCP servers for GitHub

Best forProfile
1
github/github-mcp-server
A developer with a remote-capable MCP host who needs GitHub repository, issue, and pull request context: it documents a hosted HTTP endpoint.Community favourite33,668+6543 days ago92.2
2
jgravelle/jcodemunch-mcp
Exploring a large GitHub repository where whole-file reads waste context: it indexes source once, then serves targeted symbols through seven tools.Community favourite2,779+91today84.8
3
kunwarVivek/mcp-github-project-manager
Planning sprints and roadmaps for teams that live in GitHub Projects V2: it exposes milestone, sprint, and project tools alongside issue and pull request management.Steady103+326 days ago62.9
4
bgauryy/octocode
Researching code across a local checkout and GitHub repositories: it exposes 13 tools for local ripgrep/LSP search plus GitHub code, repo, PR, and package lookup.Community favourite950+854 days ago61.9
5
mnemox-ai/idea-reality-mcp
Before starting a new project: it exposes one idea_check tool that scans GitHub, Hacker News, npm, PyPI, and Stack Overflow for existing competitors.Community favourite822+130 days ago58.7
6
kucherenko/jscpd
Before committing code to a project already scanned by jscpd, check_duplication compares a snippet against existing duplication.Community favourite6,408+193today57.1
7
Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server
Debugging with AI coding assistants: its README documents web_search and get_content returning StackOverflow answers, GitHub Issue comments, arXiv content, and Wikipedia text in one call.Community favourite399+68 days ago57.1
8
idosal/git-mcp
AI assistants needing current documentation and code from a specific GitHub repository: remote MCP URLs for repositories and GitHub Pages avoid local setup.Abandoned but popular8,461+63163 days ago55.9
9
duriantaco/skylos
AI coding agents working in Python repositories: it exposes 16 static-analysis tools covering dead code, security, secrets, quality, architecture, and remediation.Community favourite847+16today55.1
10
miantiao-me/github-stars
When you want natural language search over your own GitHub starred repositories: it indexes their metadata and README content and documents a search_github_stars MCP tool.Steady117+1yesterday54.4
11
yusufkaraaslan/Skill_Seekers
Turning a GitHub repository into a Claude skill or RAG pipeline: it exposes 40 tools for scraping, packaging, enhancing, and exporting that source.Community favourite15,143+16818 days ago52.4
12
crypto-ninja/mcp-server-for-Github
Automating GitHub Actions, pull requests, code search, file operations, and repository management through MCP: it documents code-first tools with compact, json, and markdown response formats.Steady5082 days ago49.1
13
pipelab-org/pipelock
Agents calling MCP servers with secrets in their environment: it wraps them with bidirectional scanning for credential leaks, prompt injection, and tool description poisoning.Community favourite933+40today49.1
14
Keesan12/martin-loop
AI coding-agent runs needing spend limits, loop prevention, and verification: it exposes 24 MCP tools for preflight, run control, status, dossiers, and GitHub PR creation.Community favourite412+314yesterday47.8
15
DollhouseMCP/mcp-server
Creating, activating, and sharing reusable AI personas and modular elements: it exposes five MCP tools for element lifecycle operations plus GitHub-backed community sharing.Steady46+410 days ago45.4
16
raccioly/docguard
Teams treating documentation as the source of truth before code: it exposes read-only guard, score and diagnose tools that validate canonical docs and return findings.Community favourite32+33 days ago45.0
17
spences10/mcp-omnisearch
For an MCP client that needs web search, AI answers, and URL extraction from provider APIs: it exposes them as web_search, ai_search, and web_extract.Community favourite356+92 days ago44.8
18
hetaoBackend/mcp-github-trending
Discovering what is trending on GitHub: it exposes two tools, one for trending repositories and one for trending developers, filterable by language and period.Dormant60+4214 days ago44.7
19
roboticforce/sugar
Coding agents losing project context between sessions: its MCP memory server stores and searches per-project and global memories in local SQLite.Steady98+54 days ago44.5
20
node9-ai/node9-proxy
Teams running coding agents with GitHub access: it checks tool calls and holds destructive git, SQL, and shell actions for review.Community favourite219+4today42.8

The ranking, with the evidence

Each position is a weighted mean of adoption (40%), maintenance (24%), momentum (14%), tool description quality (13%) and trust (9%), multiplied by three attenuators: how directly the server is about GitHub (named for it, declaring it, tagged with it, or merely mentioning it), whether its repository is still moving, and how much independent evidence of adoption it has. Open the score on any entry to see every number, including the ones marked ≈, which were imputed from the median of the other candidates rather than measured. The maintenance grade on each entry is mostly issue responsiveness, release recency and open security alerts rather than commits, so a recent commit beside a low grade is two different measurements rather than a contradiction.

  • Abandoned but popular: People use it, but its default branch has stopped moving. Fine to keep running, risky to adopt.
  • Community favourite: Widely adopted and still actively maintained.
  • Dormant: Neither changing nor widely adopted. Here because it still matches the search.
  • Emerging: Small audience, growing quickly, maintained. The bet with the most upside.
  • Steady: Maintained, modest audience, no surprises in either direction.
  • 1

    GitHub MCP Server

    Community favouriteNo change
    github/github-mcp-server ↗

    Best for: A developer with a remote-capable MCP host who needs GitHub repository, issue, and pull request context: it documents a hosted HTTP endpoint.

    GitHub MCP Server connects AI tools to GitHub through a hosted remote MCP endpoint, with installation guides for VS Code, Claude, Codex, Cursor, Windsurf, Zed, and other hosts, and supports OAuth or a GitHub Personal Access Token. Before choosing it, note that the remote server requires a compatible MCP host with remote support and applicable policies enabled, and OAuth access assumes the host application has its own GitHub App or OAuth App configuration.

    GitHub stars
    33,668
    Stars over the last 90 days: 31,710 at the earliest of 39 readings and 33,843 at the latest, up 2,133.
    Stars / 30 days
    +654
    npm / typical week
    Ships no npm package
    PyPI / typical week
    no attributed package
    Tools exposed
    never inspected
    Last commit
    3 days ago
    Commits / 12 weeks
    171
    Weekly commits over the last 12 weeks, oldest first: 5, 15, 59, 22, 11, 14, 2, 0, 1, 31, 4, 3. 167 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    Not graded
    Score 92.2: show every number behind it
    • Adoption100 / 100 · weight 40%
      • GitHub stars100
      • npm downloadsnot measured
        no npm package
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum74 / 100 · weight 14%
      • Stars gained, relative to size57
      • Stars gained, absolute100
      • npm download trendnot measured
        no download history for the selected registry
    • Tool quality≈68 / 100 · weight 13%
      • Tool description quality≈68
        tool descriptions not yet scored
      • Built and inspected by Glamanot measured
        never built and inspected by Glama
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integrates100
    Weighted mean of the five
    92.2
    × relevance: the keyword is dedicated here
    1.00
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    92.2
  • 2

    github-codemunch-mcp

    Community favouriteNo change
    jgravelle/jcodemunch-mcp ↗

    Best for: Exploring a large GitHub repository where whole-file reads waste context: it indexes source once, then serves targeted symbols through seven tools.

    It parses repository source with tree-sitter and exposes 7 tools: index_repo builds a local index, and list_repos, get_file_tree, get_file_outline, get_symbol, get_symbols, and search_symbols query it. Indexing is a prerequisite, since the retrieval tools only cover repositories already indexed into local storage, and the license is free for personal use while commercial use requires a paid license.

    GitHub stars
    2,779
    Stars over the last 90 days: 2,192 at the earliest of 53 readings and 2,785 at the latest, up 593.
    Stars / 30 days
    +91
    npm / typical week
    Ships no npm package
    PyPI / typical week
    10.8K
    jcodemunch-mcp
    Tools exposed
    7
    Last commit
    today
    Commits / 12 weeks
    1,149
    Weekly commits over the last 12 weeks, oldest first: 76, 86, 89, 116, 119, 235, 100, 49, 61, 54, 51, 105. 1,141 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    A
    Score 84.8: show every number behind it
    • Adoption99 / 100 · weight 40%
      • GitHub stars86
      • PyPI downloads86
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum82 / 100 · weight 14%
      • Stars gained, relative to size73
      • Stars gained, absolute79
      • PyPI download trend100
    • Tool quality76 / 100 · weight 13%
      • Tool description quality68
      • Built and inspected by Glama100
    • Trust0 / 100 · weight 9%
      • License0
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    84.8
    × relevance: the keyword is dedicated here
    1.00
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    84.8
  • 3

    mcp-github-project-manager

    Steady▲ 1
    kunwarVivek/mcp-github-project-manager ↗

    Best for: Planning sprints and roadmaps for teams that live in GitHub Projects V2: it exposes milestone, sprint, and project tools alongside issue and pull request management.

    The server exposes 84 tools for creating and updating GitHub projects, milestones, issues, draft issues, pull requests, reviews, and sprints, and its README documents MCP resources, prompts, and stdio setup for Claude, Codex, Cursor, Windsurf, and Roo. Choosing it assumes the work already lives in GitHub Projects V2, since all state is kept in GitHub issues, project fields, and comments rather than an external database.

    GitHub stars
    103
    Stars over the last 90 days: 94 at the earliest of 86 readings and 104 at the latest, up 10.
    Stars / 30 days
    +3
    npm / typical week
    49
    PyPI / typical week
    no attributed package
    Tools exposed
    84
    Last commit
    26 days ago
    Commits / 12 weeks
    56
    Weekly commits over the last 12 weeks, oldest first: 7, 31, 0, 0, 1, 4, 0, 13, 0, 0, 0, 5. 61 in total, and the last figure is the current week so far.
    Maintenance grade
    C
    Tool descriptions
    C
    Score 62.9: show every number behind it
    • Adoption56 / 100 · weight 40%
      • GitHub stars50
      • npm downloads36
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance89 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade55
      • Commit cadence100
    • Momentum27 / 100 · weight 14%
      • Stars gained, relative to size44
      • Stars gained, absolute24
      • npm download trend0
    • Tool quality51 / 100 · weight 13%
      • Tool description quality43
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    62.9
    × relevance: the keyword is dedicated here
    1.00
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    62.9
  • 4

    Octocode MCP

    Community favourite▼ 1
    bgauryy/octocode ↗

    Best for: Researching code across a local checkout and GitHub repositories: it exposes 13 tools for local ripgrep/LSP search plus GitHub code, repo, PR, and package lookup.

    Octocode MCP exposes 13 tools for code research spanning local ripgrep and LSP navigation plus GitHub code search, file reads, repository structure, repository and pull request search, and NPM/Python package lookup. Before choosing it, note that running the MCP server through npx requires Node.js 20.12+, and GitHub authentication is optional but unlocks private repositories and higher API rate limits.

    GitHub stars
    950
    Stars over the last 90 days: 899 at the earliest of 25 readings and 953 at the latest, up 54.
    Stars / 30 days
    +8
    npm / typical week
    2.1K
    PyPI / typical week
    no attributed package
    Tools exposed
    13
    Last commit
    54 days ago
    Commits / 12 weeks
    16
    Weekly commits over the last 12 weeks, oldest first: 6, 7, 2, 1, 0, 0, 0, 0, 0, 0, 0, 0. 16 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    A
    Score 61.9: show every number behind it
    • Adoption80 / 100 · weight 40%
      • GitHub stars74
      • npm downloads35
        downloads show none of the weekday rhythm human traffic has; halved
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance90 / 100 · weight 24%
      • Last commit touching this server88
        dated from the last commit on the default branch, re-read from GitHub at publication; github.com shows a push 1 days ago, which counts every ref; the stored date would have published 1 days ago
      • Repository maintenance grade100
      • Commit cadence85
    • Momentum27 / 100 · weight 14%
      • Stars gained, relative to size35
      • Stars gained, absolute38
      • npm download trend0
    • Tool quality100 / 100 · weight 13%
      • Tool description quality93
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    79.4
    × relevance: the keyword is declared here
    0.78
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    61.9
  • 5

    idea-reality-mcp

    Community favourite▲ 1
    mnemox-ai/idea-reality-mcp ↗

    Best for: Before starting a new project: it exposes one idea_check tool that scans GitHub, Hacker News, npm, PyPI, and Stack Overflow for existing competitors.

    The server exposes one tool, idea_check, which takes a plain-English idea description and returns a reality signal with competitor data, trend direction, and pivot suggestions gathered from GitHub, Hacker News, npm, PyPI, and Stack Overflow. It does not cover GitHub repository operations, and local installation assumes a uvx-capable Python environment (the README also documents a hosted REST endpoint and a Smithery remote install).

    GitHub stars
    822
    Stars over the last 90 days: 767 at the earliest of 11 readings and 825 at the latest, up 58.
    Stars / 30 days
    +1
    npm / typical week
    Ships no npm package
    PyPI / typical week
    70
    idea-reality-mcp
    Tools exposed
    1
    Last commit
    30 days ago
    Commits / 12 weeks
    5
    Weekly commits over the last 12 weeks, oldest first: 2, 0, 2, 0, 0, 0, 0, 1, 0, 0, 0, 0. 5 in total, and the last figure is the current week so far.
    Maintenance grade
    C
    Tool descriptions
    A
    Score 58.7: show every number behind it
    • Adoption76 / 100 · weight 40%
      • GitHub stars73
      • PyPI downloads20
        PyPI downloads show no weekday rhythm; halved
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance82 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade55
      • Commit cadence65
    • Momentum30 / 100 · weight 14%
      • Stars gained, relative to size16
      • Stars gained, absolute16
      • PyPI download trend72
    • Tool quality93 / 100 · weight 13%
      • Tool description quality85
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    75.3
    × relevance: the keyword is declared here
    0.78
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    58.7
  • 6

    jscpd

    Community favourite▲ 1
    kucherenko/jscpd ↗

    Best for: Before committing code to a project already scanned by jscpd, check_duplication compares a snippet against existing duplication.

    jscpd is a duplicate code detector whose built-in MCP server, started with jscpd --mcp, exposes four tools over stdio: check_duplication, get_file_clones, get_statistics, and check_current_directory. Before choosing it, know that it checks against a scanned project: check_duplication compares snippets to the last scan, while check_current_directory re-scans only the paths the server was started with.

    GitHub stars
    6,408
    Stars over the last 90 days: 6,067 at the earliest of 30 readings and 6,419 at the latest, up 352.
    Stars / 30 days
    +193
    npm / typical week
    Ships no npm package
    PyPI / typical week
    no attributed package
    Tools exposed
    4
    Last commit
    today
    Commits / 12 weeks
    671
    Weekly commits over the last 12 weeks, oldest first: 12, 10, 78, 45, 42, 48, 65, 76, 65, 91, 108, 46. 686 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    A
    Score 57.1: show every number behind it
    • Adoption95 / 100 · weight 40%
      • GitHub stars95
      • npm downloadsnot measured
        no npm package
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum79 / 100 · weight 14%
      • Stars gained, relative to size71
      • Stars gained, absolute92
      • npm download trendnot measured
        no download history for the selected registry
    • Tool quality100 / 100 · weight 13%
      • Tool description quality93
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integrates100
    Weighted mean of the five
    95.2
    × relevance: the keyword is tagged here
    0.60
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    57.1
  • 7

    Kindly Web Search MCP Server

    Community favourite▼ 2
    Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server ↗

    Best for: Debugging with AI coding assistants: its README documents web_search and get_content returning StackOverflow answers, GitHub Issue comments, arXiv content, and Wikipedia text in one call.

    The README documents web_search for top results with title, link, snippet, and page_content, and get_content for page content in Markdown. The README positions it within a suite that also calls for installing Serena, Lad MCP Server, and tdd separately.

    GitHub stars
    399
    Stars over the last 90 days: 363 at the earliest of 18 readings and 399 at the latest, up 36.
    Stars / 30 days
    +6
    npm / typical week
    Ships no npm package
    PyPI / typical week
    no attributed package
    Tools exposed
    never inspected
    Last commit
    8 days ago
    Commits / 12 weeks
    306
    Weekly commits over the last 12 weeks, oldest first: 16, 0, 0, 4, 0, 158, 96, 26, 0, 5, 1, 0. 306 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    Not graded
    Score 57.1: show every number behind it
    • Adoption65 / 100 · weight 40%
      • GitHub stars65
      • npm downloadsnot measured
        no npm package
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum38 / 100 · weight 14%
      • Stars gained, relative to size42
      • Stars gained, absolute33
      • npm download trendnot measured
        no download history for the selected registry
    • Tool quality≈68 / 100 · weight 13%
      • Tool description quality≈68
        tool descriptions not yet scored
      • Built and inspected by Glamanot measured
        never built and inspected by Glama
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    73.2
    × relevance: the keyword is declared here
    0.78
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    57.1
  • 8

    GitMCP

    Abandoned but popularNo change
    idosal/git-mcp ↗

    Best for: AI assistants needing current documentation and code from a specific GitHub repository: remote MCP URLs for repositories and GitHub Pages avoid local setup.

    GitMCP exposes 32 tools for Git operations, GitHub and Gitea repository management, authentication, configuration, releases, backups, and health checks, and its README documents remote MCP endpoints for connecting AI assistants to GitHub repository or GitHub Pages documentation. Before choosing it, note that the README's connection instructions assume an MCP-capable client, with Claude Desktop requiring the mcp-remote npm package.

    GitHub stars
    8,461
    Stars over the last 90 days: 8,308 at the earliest of 11 readings and 8,474 at the latest, up 166.
    Stars / 30 days
    +63
    npm / typical week
    downloads not counted
    PyPI / typical week
    no attributed package
    Tools exposed
    32
    Last commit
    163 days ago
    Commits / 12 weeks
    0
    Maintenance grade
    F
    Tool descriptions
    C
    Score 55.9: show every number behind it
    • Adoption98 / 100 · weight 40%
      • GitHub stars98
      • npm downloadsnot measured
        npm names no repository for git-mcp, so its downloads cannot be attributed
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance40 / 100 · weight 24%
      • Last commit touching this server68
      • Repository maintenance grade5
      • Commit cadence5
    • Momentum50 / 100 · weight 14%
      • Stars gained, relative to size35
      • Stars gained, absolute73
      • npm download trendnot measured
        no download history for the selected registry
    • Tool quality53 / 100 · weight 13%
      • Tool description quality45
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    71.7
    × relevance: the keyword is declared here
    0.78
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    55.9
  • 9

    mcp-skylos

    Community favouriteNo change
    duriantaco/skylos ↗

    Best for: AI coding agents working in Python repositories: it exposes 16 static-analysis tools covering dead code, security, secrets, quality, architecture, and remediation.

    mcp-skylos exposes 16 tools, including analyze for dead code, security_scan, quality_check, secrets_scan, architecture_check, and remediate, returning findings with file paths, line numbers, and severity. Before choosing it, note that Python has the deepest coverage and Go dead-code and security checks require the native skylos-go engine, so other language checks are narrower.

    GitHub stars
    847
    Stars over the last 90 days: 479 at the earliest of 46 readings and 848 at the latest, up 369.
    Stars / 30 days
    +16
    npm / typical week
    Ships no npm package
    PyPI / typical week
    15.5K
    skylos
    Tools exposed
    16
    Last commit
    today
    Commits / 12 weeks
    266
    Weekly commits over the last 12 weeks, oldest first: 22, 17, 14, 14, 8, 14, 31, 22, 16, 35, 39, 41. 273 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    A
    Score 55.1: show every number behind it
    • Adoption100 / 100 · weight 40%
      • GitHub stars73
      • PyPI downloads89
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum60 / 100 · weight 14%
      • Stars gained, relative to size52
      • Stars gained, absolute49
      • PyPI download trend89
    • Tool quality81 / 100 · weight 13%
      • Tool description quality73
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    91.8
    × relevance: the keyword is tagged here
    0.60
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    55.1
  • 10

    GitHub Stars MCP Server

    SteadyNo change
    miantiao-me/github-stars ↗

    Best for: When you want natural language search over your own GitHub starred repositories: it indexes their metadata and README content and documents a search_github_stars MCP tool.

    The server fetches a user's GitHub starred repositories, processes their README content, stores the data in Cloudflare R2, builds a searchable index with AutoRAG, and documents a search_github_stars MCP tool for natural language queries. Before choosing it, note that setup requires a GitHub Personal Access Token with repo scope, a Cloudflare account with R2 and AutoRAG configured, and API key authentication for the MCP server.

    GitHub stars
    117
    Stars over the last 90 days: 112 at the earliest of 15 readings and 117 at the latest, up 5.
    Stars / 30 days
    +1
    npm / typical week
    downloads not counted
    PyPI / typical week
    no attributed package
    Tools exposed
    never inspected
    Last commit
    yesterday
    Commits / 12 weeks
    14
    Weekly commits over the last 12 weeks, oldest first: 1, 1, 1, 1, 1, 1, 1, 1, 1, 3, 1, 1. 14 in total, and the last figure is the current week so far.
    Maintenance grade
    B
    Tool descriptions
    Not graded
    Score 54.4: show every number behind it
    • Adoption52 / 100 · weight 40%
      • GitHub stars52
      • npm downloadsnot measured
        npm names no repository for mcp-github-stars, so its downloads cannot be attributed
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance92 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade80
      • Commit cadence85
    • Momentum24 / 100 · weight 14%
      • Stars gained, relative to size30
      • Stars gained, absolute16
      • npm download trendnot measured
        no download history for the selected registry
    • Tool quality≈68 / 100 · weight 13%
      • Tool description quality≈68
        tool descriptions not yet scored
      • Built and inspected by Glamanot measured
        never built and inspected by Glama
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    64.0
    × relevance: the keyword is dedicated here
    1.00
    × continuity: actively changing
    1.00
    × evidence: modest but real audience
    0.85
    Composite score
    54.4
  • 11

    Skill Seekers

    Community favouriteNo change
    yusufkaraaslan/Skill_Seekers ↗

    Best for: Turning a GitHub repository into a Claude skill or RAG pipeline: it exposes 40 tools for scraping, packaging, enhancing, and exporting that source.

    Skill Seekers exposes 40 MCP tools for scraping documentation sites, GitHub repositories, PDFs, videos, and local codebases, then packaging, enhancing, uploading, and exporting the results as AI skills or vector database formats. It is installed from the skill-seekers PyPI package with the mcp extra, while video scraping and platform uploads require additional extras or platform-specific API keys.

    GitHub stars
    15,143
    Stars over the last 90 days: 14,590 at the earliest of 15 readings and 15,193 at the latest, up 603.
    Stars / 30 days
    +168
    npm / typical week
    Ships no npm package
    PyPI / typical week
    1.8K
    skill-seekers
    Tools exposed
    40
    Last commit
    18 days ago
    Commits / 12 weeks
    93
    Weekly commits over the last 12 weeks, oldest first: 25, 1, 0, 1, 18, 0, 31, 13, 0, 4, 0, 0. 93 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    B
    Score 52.4: show every number behind it
    • Adoption100 / 100 · weight 40%
      • GitHub stars100
      • PyPI downloads35
        PyPI downloads show no weekday rhythm; halved
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum46 / 100 · weight 14%
      • Stars gained, relative to size43
      • Stars gained, absolute90
      • PyPI download trend0
    • Tool quality61 / 100 · weight 13%
      • Tool description quality53
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    87.3
    × relevance: the keyword is tagged here
    0.60
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    52.4
  • 12

    GitHub MCP Server

    Steady▲ 1
    crypto-ninja/mcp-server-for-Github ↗

    Best for: Automating GitHub Actions, pull requests, code search, file operations, and repository management through MCP: it documents code-first tools with compact, json, and markdown response formats.

    The server documents GitHub workflow automation for Actions monitoring, pull request management, code search, file operations, and repository management, exposing tools through an MCP interface and offering compact, json, and markdown response formats. Before choosing it, note that it is distributed on PyPI as github-mcp-server, requires Python 3.10 or newer, and has no npm package published.

    GitHub stars
    5
    Stars over the last 90 days: 4 at the earliest of 11 readings and 5 at the latest, up 1.
    Stars / 30 days
    0
    npm / typical week
    Ships no npm package
    PyPI / typical week
    98
    github-mcp-server
    Tools exposed
    never inspected
    Last commit
    82 days ago
    Commits / 12 weeks
    1
    Weekly commits over the last 12 weeks, oldest first: 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0. 1 in total, and the last figure is the current week so far.
    Maintenance grade
    B
    Tool descriptions
    Not graded
    Score 49.1: show every number behind it
    • Adoption45 / 100 · weight 40%
      • GitHub stars19
      • PyPI downloads42
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance76 / 100 · weight 24%
      • Last commit touching this server88
      • Repository maintenance grade80
      • Commit cadence40
    • Momentum25 / 100 · weight 14%
      • Stars gained, relative to size0
      • Stars gained, absolute0
      • PyPI download trend100
    • Tool quality≈68 / 100 · weight 13%
      • Tool description quality≈68
        tool descriptions not yet scored
      • Built and inspected by Glamanot measured
        never built and inspected by Glama
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    57.8
    × relevance: the keyword is dedicated here
    1.00
    × continuity: actively changing
    1.00
    × evidence: modest but real audience
    0.85
    Composite score
    49.1
  • 13

    pipelock

    Community favourite▼ 1
    pipelab-org/pipelock ↗

    Best for: Agents calling MCP servers with secrets in their environment: it wraps them with bidirectional scanning for credential leaks, prompt injection, and tool description poisoning.

    Pipelock wraps MCP servers and scans traffic in both directions for credential leaks, prompt injection, and tool description poisoning, and it documents an HTTP fetch proxy for mediated egress. CONNECT tunnel contents are inspected only when TLS interception is enabled; otherwise CONNECT traffic is scanned at the hostname and URL level.

    GitHub stars
    933
    Stars over the last 90 days: 781 at the earliest of 54 readings and 938 at the latest, up 157.
    Stars / 30 days
    +40
    npm / typical week
    Ships no npm package
    PyPI / typical week
    no attributed package
    Tools exposed
    never inspected
    Last commit
    today
    Commits / 12 weeks
    791
    Weekly commits over the last 12 weeks, oldest first: 45, 54, 66, 55, 78, 72, 67, 71, 71, 85, 89, 42. 795 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    Not graded
    Score 49.1: show every number behind it
    • Adoption74 / 100 · weight 40%
      • GitHub stars74
      • npm downloadsnot measured
        no npm package
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum74 / 100 · weight 14%
      • Stars gained, relative to size79
      • Stars gained, absolute65
      • npm download trendnot measured
        no download history for the selected registry
    • Tool quality≈68 / 100 · weight 13%
      • Tool description quality≈68
        tool descriptions not yet scored
      • Built and inspected by Glamanot measured
        never built and inspected by Glama
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    81.8
    × relevance: the keyword is tagged here
    0.60
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    49.1
  • 14

    MartinLoop — Stop Runaway Loops, Bad Code & Token Waste

    Community favouriteNo change
    Keesan12/martin-loop ↗

    Best for: AI coding-agent runs needing spend limits, loop prevention, and verification: it exposes 24 MCP tools for preflight, run control, status, dossiers, and GitHub PR creation.

    MartinLoop governs AI coding-agent runs through 24 MCP tools for planning, preflight, budget and status checks, pause/cancel/continue control, verification evidence, dossiers, evaluation, and GitHub pull-request creation. It requires the @martinloop/mcp MCP package for host installation and the martin-loop npm package for CLI use.

    GitHub stars
    412
    Stars over the last 90 days: 39 at the earliest of 32 readings and 414 at the latest, up 375.
    Stars / 30 days
    +314
    npm / typical week
    200
    PyPI / typical week
    no attributed package
    Tools exposed
    24
    Last commit
    yesterday
    Commits / 12 weeks
    447
    Weekly commits over the last 12 weeks, oldest first: 2, 0, 50, 147, 21, 65, 44, 14, 72, 5, 7, 5. 432 in total, and the last figure is the current week so far.
    Maintenance grade
    B
    Tool descriptions
    A
    Score 47.8: show every number behind it
    • Adoption69 / 100 · weight 40%
      • GitHub stars65
      • npm downloads24
        downloads show none of the weekday rhythm human traffic has; halved
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance95 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade80
      • Commit cadence100
    • Momentum75 / 100 · weight 14%
      • Stars gained, relative to size100
      • Stars gained, absolute100
      • npm download trend0
    • Tool quality76 / 100 · weight 13%
      • Tool description quality68
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    79.7
    × relevance: the keyword is tagged here
    0.60
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    47.8
  • 15

    DollhouseMCP

    SteadyNo change
    DollhouseMCP/mcp-server ↗

    Best for: Creating, activating, and sharing reusable AI personas and modular elements: it exposes five MCP tools for element lifecycle operations plus GitHub-backed community sharing.

    DollhouseMCP exposes five MCP tools split into create, read, update, delete, and execute operations for persona, skill, template, agent, memory, and ensemble elements, with an npx web setup wizard documented for MCP-compatible clients. Before choosing it, note that it assumes an MCP-compatible client and that permission hook support varies by client, with full native support listed only for Claude Code.

    GitHub stars
    46
    Stars over the last 90 days: 36 at the earliest of 35 readings and 46 at the latest, up 10.
    Stars / 30 days
    +4
    npm / typical week
    263
    PyPI / typical week
    no attributed package
    Tools exposed
    5
    Last commit
    10 days ago
    Commits / 12 weeks
    148
    Weekly commits over the last 12 weeks, oldest first: 8, 59, 0, 0, 27, 11, 0, 0, 23, 0, 20, 0. 148 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    A
    Score 45.4: show every number behind it
    • Adoption46 / 100 · weight 40%
      • GitHub stars42
      • npm downloads26
        downloads show none of the weekday rhythm human traffic has; halved
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum37 / 100 · weight 14%
      • Stars gained, relative to size62
      • Stars gained, absolute30
      • npm download trend0
    • Tool quality93 / 100 · weight 13%
      • Tool description quality85
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    68.5
    × relevance: the keyword is declared here
    0.78
    × continuity: actively changing
    1.00
    × evidence: modest but real audience
    0.85
    Composite score
    45.4
  • 16

    DocGuard

    Community favourite▲ 4
    raccioly/docguard ↗

    Best for: Teams treating documentation as the source of truth before code: it exposes read-only guard, score and diagnose tools that validate canonical docs and return findings.

    DocGuard is an npm-published MCP server (package docguard-cli, started with docguard mcp) whose 12 read-only tools report on a project's canonical documentation: docguard_guard runs the enabled validators and returns each finding with a stable code, docguard_score computes a CDD maturity grade, docguard_diagnose returns only failing and warning validators with suggested fixes, and docguard_verify_claims extracts documented numbers, limits and enums as a verification task list. Every tool is read-only, so it surfaces findings, scores and commit-stamped evidence bundles rather than editing the documentation it checks, and its evidence-verification path evaluates a .docguard-evidence.json file against bounded local sources.

    GitHub stars
    32
    Stars over the last 90 days: 22 at the earliest of 23 readings and 32 at the latest, up 10.
    Stars / 30 days
    +3
    npm / typical week
    564
    PyPI / typical week
    642
    docguard-cli
    Tools exposed
    12
    Last commit
    3 days ago
    Commits / 12 weeks
    229
    Weekly commits over the last 12 weeks, oldest first: 0, 0, 1, 0, 8, 2, 69, 34, 13, 46, 55, 5. 233 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    A
    Score 45.0: show every number behind it
    • Adoption65 / 100 · weight 40%
      • GitHub stars38
      • PyPI downloads60
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum40 / 100 · weight 14%
      • Stars gained, relative to size51
      • Stars gained, absolute24
      • PyPI download trendnot measured
        no download history for the selected registry
    • Tool quality78 / 100 · weight 13%
      • Tool description quality70
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    75.0
    × relevance: the keyword is tagged here
    0.60
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    45.0
  • 17

    mcp-omnisearch

    Community favouriteNo change
    spences10/mcp-omnisearch ↗

    Best for: For an MCP client that needs web search, AI answers, and URL extraction from provider APIs: it exposes them as web_search, ai_search, and web_extract.

    The server exposes web_search, ai_search, and web_extract tools for web search, sourced AI answers, and content extraction or summarization. It relies on provider API keys supplied through environment variables, and providers without keys are skipped while the rest keep working.

    GitHub stars
    356
    Stars over the last 90 days: 335 at the earliest of 38 readings and 356 at the latest, up 21.
    Stars / 30 days
    +9
    npm / typical week
    downloads not counted
    PyPI / typical week
    no attributed package
    Tools exposed
    3
    Last commit
    2 days ago
    Commits / 12 weeks
    117
    Weekly commits over the last 12 weeks, oldest first: 6, 4, 17, 7, 17, 10, 1, 19, 6, 11, 14, 7. 119 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    A
    Score 44.8: show every number behind it
    • Adoption64 / 100 · weight 40%
      • GitHub stars64
      • npm downloadsnot measured
        npm names no repository for mcp-omnisearch, so its downloads cannot be attributed
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum48 / 100 · weight 14%
      • Stars gained, relative to size53
      • Stars gained, absolute40
      • npm download trendnot measured
        no download history for the selected registry
    • Tool quality73 / 100 · weight 13%
      • Tool description quality65
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    74.7
    × relevance: the keyword is tagged here
    0.60
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    44.8
  • 18

    mcp-github-trending

    Dormant▼ 2
    hetaoBackend/mcp-github-trending ↗

    Best for: Discovering what is trending on GitHub: it exposes two tools, one for trending repositories and one for trending developers, filterable by language and period.

    mcp-github-trending exposes two tools, get_github_trending_repositories and get_github_trending_developers, returning JSON records for each entry with optional filters for programming language, spoken language, and a daily, weekly, or monthly window. Its scope stops at the trending lists themselves: there is no repository search, issue, or pull request tooling, the README documents no authentication token for reaching private repositories, and the stated setup assumes Python 3.12 with a local stdio launch through uv or uvx.

    GitHub stars
    60
    Stars over the last 90 days: 53 at the earliest of 11 readings and 61 at the latest, up 8.
    Stars / 30 days
    +4
    npm / typical week
    Ships no npm package
    PyPI / typical week
    no attributed package
    Tools exposed
    2
    Last commit
    214 days ago
    Commits / 12 weeks
    0
    Maintenance grade
    D
    Tool descriptions
    B
    Score 44.7: show every number behind it
    • Adoption45 / 100 · weight 40%
      • GitHub stars45
      • npm downloadsnot measured
        no npm package
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance46 / 100 · weight 24%
      • Last commit touching this server68
      • Repository maintenance grade30
      • Commit cadence5
    • Momentum47 / 100 · weight 14%
      • Stars gained, relative to size59
      • Stars gained, absolute29
      • npm download trendnot measured
        no download history for the selected registry
    • Tool quality63 / 100 · weight 13%
      • Tool description quality55
      • Built and inspected by Glama100
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    52.6
    × relevance: the keyword is dedicated here
    1.00
    × continuity: actively changing
    1.00
    × evidence: modest but real audience
    0.85
    Composite score
    44.7
  • 19

    Sugar

    Steady▼ 1
    roboticforce/sugar ↗

    Best for: Coding agents losing project context between sessions: its MCP memory server stores and searches per-project and global memories in local SQLite.

    Sugar documents MCP entry points sugar mcp memory and optional sugar mcp tasks, with agent calls such as store_learning and search_memories against per-project and global SQLite memory stores. The dependency to note is that memory setup requires installing the sugarai PyPI package via pipx and running sugar init, with local SQLite storage and no API keys.

    GitHub stars
    98
    Stars over the last 90 days: 92 at the earliest of 13 readings and 98 at the latest, up 6.
    Stars / 30 days
    +5
    npm / typical week
    Ships no npm package
    PyPI / typical week
    116
    sugarai
    Tools exposed
    never inspected
    Last commit
    4 days ago
    Commits / 12 weeks
    12
    Weekly commits over the last 12 weeks, oldest first: 0, 0, 0, 1, 0, 0, 0, 9, 0, 0, 2, 4. 16 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    Not graded
    Score 44.5: show every number behind it
    • Adoption53 / 100 · weight 40%
      • GitHub stars50
      • PyPI downloads22
        PyPI downloads show no weekday rhythm; halved
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance97 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence85
    • Momentum34 / 100 · weight 14%
      • Stars gained, relative to size56
      • Stars gained, absolute30
      • PyPI download trend0
    • Tool quality≈68 / 100 · weight 13%
      • Tool description quality≈68
        tool descriptions not yet scored
      • Built and inspected by Glamanot measured
        never built and inspected by Glama
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    67.1
    × relevance: the keyword is declared here
    0.78
    × continuity: actively changing
    1.00
    × evidence: modest but real audience
    0.85
    Composite score
    44.5
  • 20

    Node9-Proxy

    Community favourite▼ 1
    node9-ai/node9-proxy ↗

    Best for: Teams running coding agents with GitHub access: it checks tool calls and holds destructive git, SQL, and shell actions for review.

    Node9-Proxy runs as an MCP server over stdio (npx -y node9-ai mcp-server) so an agent can read its status, posture, audit record, and policy and explain any decision, while it also checks tool calls and holds destructive git, SQL, and shell actions for review. Before choosing it, note that it requires Node.js 22+.

    GitHub stars
    219
    Stars over the last 90 days: 210 at the earliest of 37 readings and 219 at the latest, up 9.
    Stars / 30 days
    +4
    npm / typical week
    1.6K
    PyPI / typical week
    no attributed package
    Tools exposed
    never inspected
    Last commit
    today
    Commits / 12 weeks
    723
    Weekly commits over the last 12 weeks, oldest first: 0, 3, 24, 33, 75, 140, 68, 82, 138, 21, 75, 108. 767 in total, and the last figure is the current week so far.
    Maintenance grade
    A
    Tool descriptions
    Not graded
    Score 42.8: show every number behind it
    • Adoption64 / 100 · weight 40%
      • GitHub stars59
      • npm downloads34
        downloads show none of the weekday rhythm human traffic has; halved
      • Used through Glamanot measured
        not used through Glama in the last 30 days
    • Maintenance100 / 100 · weight 24%
      • Last commit touching this server100
      • Repository maintenance grade100
      • Commit cadence100
    • Momentum29 / 100 · weight 14%
      • Stars gained, relative to size44
      • Stars gained, absolute29
      • npm download trend0
    • Tool quality≈68 / 100 · weight 13%
      • Tool description quality≈68
        tool descriptions not yet scored
      • Built and inspected by Glamanot measured
        never built and inspected by Glama
    • Trust100 / 100 · weight 9%
      • License100
      • Published by the vendor it integratesnot measured
        not published by the vendor it integrates
    Weighted mean of the five
    71.3
    × relevance: the keyword is tagged here
    0.60
    × continuity: actively changing
    1.00
    × evidence: widely adopted
    1.00
    Composite score
    42.8

Questions people ask

How do I choose between GitHub MCP Server (github/github-mcp-server) and github-codemunch-mcp (jgravelle/jcodemunch-mcp)?

Choose GitHub MCP Server (github/github-mcp-server) when you need repository, issue, and pull request context through a hosted HTTP endpoint. It had 33,668 stars, 654 stars gained per 30 days, a commit 3 days ago, and 171 commits in the last 12 weeks. Choose github-codemunch-mcp (jgravelle/jcodemunch-mcp) when whole-file reads waste context: it exposes 7 tools, its tool descriptions are graded A, it had 10,808 PyPI downloads in a typical week, and its default branch had 1,149 commits in the last 12 weeks.

What does GitHub MCP Server (github/github-mcp-server) require before it will run?

It is for a developer with a remote-capable MCP host, and it documents a hosted HTTP endpoint. It has no npm package published and no PyPI package attributed. Its default branch had a commit 3 days ago, it had 171 commits in the last 12 weeks, and it is not archived.

What does github-codemunch-mcp (jgravelle/jcodemunch-mcp) not cover?

Its described scope is exploring a GitHub repository with tree-sitter AST parsing and serving targeted symbols through 7 tools. It does not document GitHub Projects V2, milestones, or sprints. For those, mcp-github-project-manager (kunwarVivek/mcp-github-project-manager) exposes milestone, sprint, and project tools alongside issue and pull request management.

What should I be careful about with GitMCP (idosal/git-mcp)?

GitMCP (idosal/git-mcp) is Abandoned but popular, with 8,461 stars and 63 stars gained per 30 days. Its default branch had a commit 163 days ago and 0 commits in the last 12 weeks, though it exposes 32 tools and its tool descriptions are graded C. It is described as offering remote MCP URLs for repositories and GitHub Pages, but those commit figures show little activity.

How do I choose between GitHub MCP Server (github/github-mcp-server) and mcp-github-project-manager (kunwarVivek/mcp-github-project-manager)?

Choose GitHub MCP Server (github/github-mcp-server) for repository, issue, and pull request context over a hosted HTTP endpoint; it had 33,668 stars, 654 stars gained per 30 days, and 171 commits in the last 12 weeks. Choose mcp-github-project-manager (kunwarVivek/mcp-github-project-manager) when planning sprints and roadmaps in GitHub Projects V2: it exposes 84 tools, had 49 npm downloads in a typical week, and had 56 commits in the last 12 weeks. Its tool descriptions are graded C.

What should I use to guard an agent with GitHub access?

Node9-Proxy (node9-ai/node9-proxy) checks tool calls and holds destructive git, SQL, and shell actions for review, and it had a commit 0 days ago, 723 commits in the last 12 weeks, and 1,607 npm downloads in a typical week. pipelock (pipelab-org/pipelock) wraps any MCP server with bidirectional scanning for credential leaks, prompt injection, and tool description poisoning, and it had a commit 0 days ago and 791 commits in the last 12 weeks. Both are not archived. Choose pipelock for scanning around the MCP server, and Node9-Proxy for reviewing destructive actions before they run.

Other comparisons