20 Best Social Media MCP Servers, Compared (November 2026)
The short answer
Reach for LinkedIn MCP Server (stickerdaniel/linkedin-mcp-server) first. It covers LinkedIn profiles, companies, jobs, messaging, feed and connection requests through an authenticated browser session, and the evidence behind it is the strongest here: 4,119 GitHub stars, 347 stars gained per 30 days, 9,464 PyPI downloads in a typical week, a tool description grade of A, and a commit 1 day ago with 436 commits in the last 12 weeks. If you do not need LinkedIn and want one server that can extract data from many different sites and platforms instead, actors-mcp-server is the better answer: 14,820 GitHub stars, 20,157 npm downloads in a typical week, 13 tools, a tool description grade of A, and a commit 0 days ago.
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
- 1LinkedIn MCP Server : For researching LinkedIn people, companies, and jobs through an authenticated browser session: it exposes 17 tools for profiles, companies, jobs, messaging, feed, and connection requests.
- 2Reddit Buddy MCP : Anonymous subreddit browsing when no approved Reddit API access is available: browse_subreddit fetches posts from public RSS feeds when available.
- 3X Twitter Scraper : Teams needing X search and user lookups via hosted MCP server: it exposes explore and xquik tools for catalog lookup and approved API requests.
- 4reddit-mcp-server : Users who need to browse Reddit and publish posts or replies from an MCP client: it exposes create_post, reply_to_post, and related read tools.
- 5AdLoop: Google Ads, Reddit Ads & GA4 : Managing Google Ads and GA4 from an AI client: 153 tools for reports and changes behind previews, Google-validated dry runs, and budget and bid caps.
Which one, for your situation
| Your situation | What to use |
|---|---|
| I need LinkedIn people, companies and job postings | LinkedIn MCP Server (stickerdaniel/linkedin-mcp-server) had 4,119 GitHub stars, 347 stars gained per 30 days, 9,464 PyPI downloads in a typical week, a tool description grade of A and a commit 1 day ago, which is the deepest set of signals for LinkedIn work. |
| I have no approved Reddit API access | Reddit Buddy MCP requires no API keys and offers optional authentication for higher rate limits, with 5 tools, 874 GitHub stars and a commit 3 days ago. |
| I want one server for many sites and platforms | actors-mcp-server searches the Apify Store and calls Actors for you, with 13 tools, 14,820 GitHub stars, 20,157 npm downloads in a typical week and a commit 0 days ago. |
| I need to publish to X with human review | x-use drives a real Chrome session with exported cookies, needs no X API key, and stages writes as drafts by default, with 38 PyPI downloads in a typical week and a commit 3 days ago. |
| I need to research a brand's Facebook and Instagram ads | Facebook Ads Library MCP Server (proxy-intell/facebook-ads-library-mcp) is self-hosted, exposes 8 tools for ad retrieval and image or video analysis, and had a commit 3 days ago. |
| I need Google Ads and GA4 work with guarded writes | AdLoop puts previews, Google-validated dry runs and budget and bid caps in front of changes, and had 292 GitHub stars, 837 PyPI downloads in a typical week and a commit 1 day ago. |
Top MCP servers for Social Media
| Profile | |||||||
|---|---|---|---|---|---|---|---|
| 1 | Community favourite | 4,119 | +347 | 9.5K | yesterday | 91.7 | |
| 2 | Community favourite | 874 | +35 | 1.8K | 3 days ago | 86.5 | |
| 3 | Community favourite | 227 | +20 | 347 | today | 81.1 | |
| 4 | Community favourite | 316 | +41 | 746 | 5 days ago | 79.9 | |
| 5 | Community favourite | 292 | +17 | 837 | yesterday | 79.8 | |
| 6 | Community favourite | 7,485 | +1,283 | – | today | 75.3 | |
| 7 | Emerging | 166 | +11 | 298 | 12 days ago | 75.0 | |
| 8 | Community favourite | 14,820 | +5,059 | 20.2K | today | 74.4 | |
| 9 | Emerging | 212 | +24 | – | 12 days ago | 74.2 | |
| 10 | Community favourite | 321 | +11 | 28 | 3 days ago | 73.9 | |
| 11 | Community favourite | 407 | -1 | 277 | 2 days ago | 71.6 | |
| 12 | Community favourite | 263 | +11 | – | 16 days ago | 70.9 | |
| 13 | Community favourite | 1,438 | -1 | – | 45 days ago | 70.9 | |
| 14 | Community favourite | 6,928 | +3,248 | – | 9 days ago | 70.8 | |
| 15 | Community favourite | 283 | +16 | – | 68 days ago | 68.5 | |
| 16 | Steady | 208 | +8 | – | 78 days ago | 67.6 | |
| 17 | Community favourite | 643 | +76 | 452 | yesterday | 63.3 | |
| 18 | Abandoned but popular | 366 | +1 | – | 92 days ago | 59.8 | |
| 19 | Abandoned but popular | 234 | +6 | – | 196 days ago | 58.6 | |
| 20 | Community favourite | 180 | +11 | 38 | 3 days ago | 57.9 |
The ranking, with the evidence
- 1stickerdaniel/linkedin-mcp-server ↗
LinkedIn MCP Server
Community favouriteBest for: For researching LinkedIn people, companies, and jobs through an authenticated browser session: it exposes 17 tools for profiles, companies, jobs, messaging, feed, and connection requests.
It provides AI assistants with 17 tools to scrape LinkedIn profiles, companies, and job postings, search people and companies, read messaging conversations and the feed, and send messages or connection requests via an authenticated browser session. Before choosing it, note that it requires an authenticated LinkedIn browser session, so the automation must run with access to a logged-in browser session.
Score 91.7: show every number behind it
- Adoption97 / 100 · weight 40%
- GitHub stars90
- PyPI downloads42PyPI downloads show no weekday rhythm; halved
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum75 / 100 · weight 14%
- Stars gained, relative to size100
- Stars gained, absolute100
- PyPI download trend2
- Tool quality73 / 100 · weight 13%
- Tool description quality65
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 91.7
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 91.7
- 2karanb192/reddit-mcp-buddy ↗
Reddit Buddy MCP
Community favouriteBest for: Anonymous subreddit browsing when no approved Reddit API access is available: browse_subreddit fetches posts from public RSS feeds when available.
Reddit Buddy MCP exposes five tools for browsing subreddits, searching Reddit, fetching post details with comments, analyzing user activity, and explaining Reddit terms, and its README documents optional posting and commenting. Before choosing it, note that anonymous browsing uses Reddit's public RSS feeds when available, while search, comments, user analysis, and full metrics require approved Reddit API access.
Score 86.5: show every number behind it
- Adoption84 / 100 · weight 40%
- GitHub stars74
- npm downloads69
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance97 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence85
- Momentum70 / 100 · weight 14%
- Stars gained, relative to size76
- Stars gained, absolute63
- npm download trend69
- Tool quality83 / 100 · weight 13%
- Tool description quality75
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 86.5
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 86.5
- 3Xquik-dev/x-twitter-scraper ↗
X Twitter Scraper
Community favouriteBest for: Teams needing X search and user lookups via hosted MCP server: it exposes explore and xquik tools for catalog lookup and approved API requests.
X Twitter Scraper exposes two MCP tools, explore and xquik, and its README documents a hosted MCP server plus REST, SDK, CLI, webhook, and Apify interfaces for X search, user lookup, timelines, followers, likes, retweets, monitoring, giveaway draws, and write actions. The package stub returns setup guidance only, and live API use requires an XQUIK_API_KEY, with private reads or X account actions also requiring a connected X account.
Score 81.1: show every number behind it
- Adoption63 / 100 · weight 40%
- GitHub stars59
- npm downloads27downloads show none of the weekday rhythm human traffic has; halved
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum84 / 100 · weight 14%
- Stars gained, relative to size96
- Stars gained, absolute53
- npm download trend100
- Tool quality85 / 100 · weight 13%
- Tool description quality77
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 81.1
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 81.1
- 4jordanburke/reddit-mcp-server ↗
reddit-mcp-server
Community favouriteBest for: Users who need to browse Reddit and publish posts or replies from an MCP client: it exposes create_post, reply_to_post, and related read tools.
reddit-mcp-server exposes 12 MCP tools for reading Reddit posts, comments, subreddit and user information, and search results, and for creating posts and replies. It requires Reddit OAuth credentials plus REDDIT_USERNAME and REDDIT_PASSWORD for write tools, and Reddit's public API is scheduled to close in March 2027.
Score 79.9: show every number behind it
- Adoption67 / 100 · weight 40%
- GitHub stars63
- npm downloads31downloads show none of the weekday rhythm human traffic has; halved
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum90 / 100 · weight 14%
- Stars gained, relative to size100
- Stars gained, absolute66
- npm download trend100
- Tool quality58 / 100 · weight 13%
- Tool description quality50
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 79.9
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 79.9
- 5kLOsk/adloop ↗
AdLoop: Google Ads, Reddit Ads & GA4
Community favouriteBest for: Managing Google Ads and GA4 from an AI client: 153 tools for reports and changes behind previews, Google-validated dry runs, and budget and bid caps.
AdLoop exposes 153 tools that let an AI client read reports and make changes across Google Ads, Reddit Ads, GA4, Tag Manager, Search Console, and Merchant Center, with previews, Google-validated dry runs, and budget and bid caps before anything goes live. Self-hosting requires a Google Cloud project with Ads API access, while AdLoop Cloud is the hosted alternative.
Score 79.8: show every number behind it
- Adoption66 / 100 · weight 40%
- GitHub stars62
- PyPI downloads31PyPI downloads show no weekday rhythm; halved
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum77 / 100 · weight 14%
- Stars gained, relative to size82
- Stars gained, absolute51
- PyPI download trend100
- Tool quality73 / 100 · weight 13%
- Tool description quality65
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 79.8
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 79.8
- 6mutonby/openshorts ↗
mcp-openshorts
Community favouriteBest for: Long videos to vertical social clips: it exposes 8 tools for URL or file processing, clip retrieval, subtitle restyling, recutting and publishing.
mcp-openshorts exposes 8 tools that process a video URL, accept a file upload, report job status, retrieve completed clips, check quota, restyle subtitles, recut clips and publish clips to connected accounts. Before choosing it, note that its self-hosted MCP endpoint is available only while the user's machine is on, while the hosted endpoint is always-on.
Score 75.3: show every number behind it
- Adoption97 / 100 · weight 40%
- GitHub stars97
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum100 / 100 · weight 14%
- Stars gained, relative to size100
- Stars gained, absolute100
- npm download trendnot measuredno download history for the selected registry
- Tool quality83 / 100 · weight 13%
- Tool description quality75
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 96.5
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 75.3
Best for: Reading Reddit without API keys when Reddit allows anonymous access: it exposes six tools for search, subreddit listings, post comment trees, and public user activity.
The Reddit MCP Server exposes six tools for Reddit search, subreddit listings, post details with nested comment trees, and public user activity, and documents stdio and streamable HTTP transports. It works without API keys only where Reddit permits anonymous access; when a network blocks unauthenticated requests, optional Reddit OAuth credentials are needed.
Score 75.0: show every number behind it
- Adoption60 / 100 · weight 40%
- GitHub stars56
- PyPI downloads26PyPI downloads show no weekday rhythm; halved
- Used through Glama14
- Maintenance93 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence65
- Momentum69 / 100 · weight 14%
- Stars gained, relative to size79
- Stars gained, absolute44
- PyPI download trend83
- Tool quality78 / 100 · weight 13%
- Tool description quality70
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 75.0
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 75.0
- 8apify/apify-mcp-server ↗
actors-mcp-server
Community favouriteBest for: Extracting data from a specific site or platform without installing a dedicated scraper: it searches Apify Store and calls Actors for you.
It exposes 13 tools that search Apify Store, fetch an Actor's input schema, call Actors, and read back run status, dataset items, and key-value store records. It assumes an Apify account: the hosted endpoint signs in with OAuth, and clients without OAuth must send an Apify API token as a bearer header.
Score 74.4: show every number behind it
- Adoption100 / 100 · weight 40%
- GitHub stars100
- npm downloads92
- Used through Glamanot measurednot 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 size100
- Stars gained, absolute100
- npm download trend14
- Tool quality88 / 100 · weight 13%
- Tool description quality80
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integrates100
- Weighted mean of the five
- 95.4
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 74.4
Best for: When extracting LinkedIn people, company, and job data from a logged-in session: it exposes profile, company, and job search/detail tools via browser automation.
The server exposes seven MCP tools for retrieving LinkedIn person and company profiles, company posts, job details, and search results, plus a browser close tool. It requires an authenticated LinkedIn account and Patchright browser binaries installed locally before use, so it is not an anonymous scraping tool.
Score 74.2: show every number behind it
- Adoption58 / 100 · weight 40%
- GitHub stars58
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance83 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade80
- Commit cadence40
- Momentum83 / 100 · weight 14%
- Stars gained, relative to size100
- Stars gained, absolute57
- npm download trendnot measuredno download history for the selected registry
- Tool quality81 / 100 · weight 13%
- Tool description quality73
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 74.2
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 74.2
- 10proxy-intell/facebook-ads-library-mcp ↗
Facebook Ads Library MCP Server
Community favouriteBest for: Self-hosted competitive ad research on a brand's current Facebook and Instagram ads: it exposes platform-ID lookup, ad retrieval, and image/video analysis tools.
The server exposes eight tools for finding a Meta platform ID, retrieving a brand's currently running ads, and analyzing ad images and videos, including batch video analysis. Self-hosting requires a ScrapeCreators API key and, for video analysis, a Google Gemini API key; the README also says ChatGPT and claude.ai need a public OAuth server rather than this repo.
Score 73.9: show every number behind it
- Adoption67 / 100 · weight 40%
- GitHub stars63
- npm downloads31
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance88 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade80
- Commit cadence65
- Momentum45 / 100 · weight 14%
- Stars gained, relative to size63
- Stars gained, absolute43
- npm download trend17
- Tool quality81 / 100 · weight 13%
- Tool description quality73
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 73.9
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 73.9
- 11EnesCinr/twitter-mcp ↗
Twitter MCP Server
Community favouriteBest for: Posting tweets with up to four images and searching Twitter from Claude: it exposes post_tweet and search_tweets tools.
It exposes three tools: post_tweet for posting tweets with optional images, search_tweets for searching Twitter, and get_home_timeline for recent tweets from followed accounts. Using it requires Twitter Developer API keys, and get_home_timeline requires a paid X API tier (Basic or higher).
Score 71.6: show every number behind it
- Adoption73 / 100 · weight 40%
- GitHub stars65
- npm downloads52
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance97 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence85
- Momentum16 / 100 · weight 14%
- Stars gained, relative to size0
- Stars gained, absolute0
- npm download trend63
- Tool quality61 / 100 · weight 13%
- Tool description quality53
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 71.6
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 71.6
- 12dialog-tools/reddit-research-mcp ↗
Reddit MCP Server
Community favouriteBest for: Researching Reddit discussions across multiple communities: it exposes discovery, requirements, and execution tools for finding subreddits and fetching posts and comments with citations.
The server exposes three MCP tools for discovering Reddit resources, checking operation requirements, and executing Reddit operations, with documented capabilities for search, post fetching, batch fetching, and comment analysis. Before choosing it, note that setup examples point to the hosted MCP endpoint at https://mcp.dialog.tools/mcp, with no npm package published and no PyPI package attributed.
Score 70.9: show every number behind it
- Adoption61 / 100 · weight 40%
- GitHub stars61
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance82 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade55
- Commit cadence65
- Momentum59 / 100 · weight 14%
- Stars gained, relative to size69
- Stars gained, absolute44
- npm download trendnot measuredno download history for the selected registry
- Tool quality76 / 100 · weight 13%
- Tool description quality68
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 70.9
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 70.9
- 136551Team/opentwitter-mcp ↗
Twitter MCP Server
Community favouriteBest for: Monitoring Twitter/X accounts for follower events, deleted tweets, and KOL followers: it exposes follower-event, deleted-tweet, and KOL-follower tools by username.
It exposes 12 tools for Twitter/X profile lookup by username or ID, recent tweets, keyword and advanced search, follower and unfollower events, deleted tweets, KOL followers, article lookup, and a monitoring watchlist. It requires an OPENNEWS_TOKEN from app.newsliquid.com/mcp, and the documented Claude Code install runs the server from a local project path with uv.
Score 70.9: show every number behind it
- Adoption79 / 100 · weight 40%
- GitHub stars79
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance88 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade80
- Commit cadence65
- Momentum0 / 100 · weight 14%
- Stars gained, relative to size0
- Stars gained, absolute0
- npm download trendnot measuredno download history for the selected registry
- Tool quality71 / 100 · weight 13%
- Tool description quality63
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 70.9
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 70.9
- 14irinabuht12-oss/google-ads-meta-ads-mcp ↗
google-meta-ads-ga4-mcp
Community favouriteBest for: Agencies managing Google Ads, Meta Ads and GA4 in one chat: it documents a hosted remote MCP with OAuth login and approval-gated writes.
The server's README documents a hosted remote MCP endpoint for Google Ads, Meta Ads, GA4 and Google Search Console, using Google or Facebook OAuth login and approval-gated writes. Before choosing it, note that it requires signing in with the Google or Facebook account that owns the ads.
Score 70.8: show every number behind it
- Adoption96 / 100 · weight 40%
- GitHub stars96
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance86 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade55
- Commit cadence85
- Momentum100 / 100 · weight 14%
- Stars gained, relative to size100
- Stars gained, absolute100
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 90.8
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 70.8
- 15RamsesAguirre777/facebook-ads-library-mcp ↗
Facebook Ads Library MCP
Community favouriteBest for: Public Facebook Ad Library research: it exposes keyword search and URL scraping for ad creative and landing-page data without a Facebook account or API token.
The server exposes search_ad_library for keyword searches and scrape_ad_library_url for supplied URLs against the public Facebook Ad Library, returning parsed ad records with fields such as advertiser, landing_url, cta, and raw_markdown. It depends on a headless Chromium browser installed through Playwright, so setup includes running playwright install chromium.
Score 68.5: show every number behind it
- Adoption61 / 100 · weight 40%
- GitHub stars61
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance70 / 100 · weight 24%
- Last commit touching this server88
- Repository maintenance grade55
- Commit cadence40
- Momentum67 / 100 · weight 14%
- Stars gained, relative to size79
- Stars gained, absolute49
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 68.5
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 68.5
- 16
Best for: For an afternoon install focused on TikTok subtitles, post metrics, or keyword search: it exposes subtitle, post-detail, and search tools.
TikTok MCP exposes three tools: tiktok_get_subtitle for video subtitles, tiktok_get_post_details for post details including likes, hashtags, and publishing time, and tiktok_search for keyword or hashtag queries. Before choosing it, note that it requires a TikNeuron account and MCP API key, configured as TIKNEURON_MCP_API_KEY.
Score 67.6: show every number behind it
- Adoption58 / 100 · weight 40%
- GitHub stars58
- npm downloadsnot measurednpm names no repository for tiktok-mcp, so its downloads cannot be attributed
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance76 / 100 · weight 24%
- Last commit touching this server88
- Repository maintenance grade80
- Commit cadence40
- Momentum52 / 100 · weight 14%
- Stars gained, relative to size61
- Stars gained, absolute38
- npm download trendnot measuredno download history for the selected registry
- Tool quality76 / 100 · weight 13%
- Tool description quality68
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 67.6
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 67.6
- 17nirholas/XActions ↗
XActions MCP Server
Community favouriteBest for: Reading X profiles, timelines and search results from an AI agent without an API key: an MCP server and CLI fetch this data.
XActions provides an MCP server for AI agents alongside a CLI with profile, tweets, search and login commands and browser scripts, all operating on X/Twitter without an X API key. Search, followers and DMs require running the login command to establish an account session, and writes are held as drafts for human approval before they are sent.
Score 63.3: show every number behind it
- Adoption74 / 100 · weight 40%
- GitHub stars70
- npm downloads28downloads show none of the weekday rhythm human traffic has; halved
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance95 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade80
- Commit cadence100
- Momentum77 / 100 · weight 14%
- Stars gained, relative to size100
- Stars gained, absolute76
- npm download trend35
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 81.2
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 63.3
- 18gomarble-ai/facebook-ads-mcp-server ↗
Facebook-Ads-MCP-Server
Abandoned but popularBest for: Meta Ads reporting inside an MCP client: the README documents tools that list ad accounts and fetch campaigns, ad sets, ads, creatives, and performance insights.
It is a Python MCP server that takes a Meta user access token as a command-line argument and exposes tools for listing ad accounts and retrieving campaigns, ad sets, ads, creatives, and insights by ID, account, or campaign. Running it requires Python 3.10+ and a token with permissions such as ads_read, and the optional one-click installer will create that token through GoMarble's server for users who do not obtain one from the Meta developer portal themselves.
Score 59.8: show every number behind it
- Adoption64 / 100 · weight 40%
- GitHub stars64
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance57 / 100 · weight 24%
- Last commit touching this server88
- Repository maintenance grade30
- Commit cadence5
- Momentum19 / 100 · weight 14%
- Stars gained, relative to size22
- Stars gained, absolute16
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 59.8
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 59.8
- 19HagaiHen/facebook-mcp-server ↗
facebook-mcp-server
Abandoned but popularBest for: Social media managers automating a Facebook Page from an LLM agent: it exposes tools to create posts, reply to comments, and hide or delete comments.
The README documents Graph API tools for creating Facebook Page posts, replying to comments, hiding or deleting comments, fetching post insights, filtering negative comments by keyword, and sending direct messages. Setup requires a Facebook Page access token and Page ID in a .env file, plus running server.py through uv with the mcp[cli] and requests dependencies.
Score 58.6: show every number behind it
- Adoption59 / 100 · weight 40%
- GitHub stars59
- npm downloadsnot measuredno npm package
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance46 / 100 · weight 24%
- Last commit touching this server68
- Repository maintenance grade30
- Commit cadence5
- Momentum43 / 100 · weight 14%
- Stars gained, relative to size50
- Stars gained, absolute33
- npm download trendnot measuredno download history for the selected registry
- Tool quality≈68 / 100 · weight 13%
- Tool description quality≈68tool descriptions not yet scored
- Built and inspected by Glamanot measurednever built and inspected by Glama
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 58.6
- × relevance: the keyword is dedicated here
- 1.00
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 58.6
- 20ihuzaifashoukat/x-use ↗
x-use
Community favouriteBest for: Publishing to X with human review and no API key: it drives a real Chrome session with exported cookies and stages writes as drafts.
x-use is an MCP server for X that uses a local Chrome session and exposes 82 tools for searching tweets and profiles, reading conversations, managing accounts and proxies, and preparing or approving posts, replies, messages, and follows, plus performing and queuing engagement actions. It requires Python 3.11+, Chromium, and a local JSON export of X session cookies, and it runs against X through an authenticated browser session rather than official API credentials.
Score 57.9: show every number behind it
- Adoption62 / 100 · weight 40%
- GitHub stars56
- PyPI downloads34
- Used through Glamanot measurednot used through Glama in the last 30 days
- Maintenance100 / 100 · weight 24%
- Last commit touching this server100
- Repository maintenance grade100
- Commit cadence100
- Momentum62 / 100 · weight 14%
- Stars gained, relative to size75
- Stars gained, absolute43
- PyPI download trendnot measuredno download history for the selected registry
- Tool quality61 / 100 · weight 13%
- Tool description quality53
- Built and inspected by Glama100
- Trust100 / 100 · weight 9%
- License100
- Published by the vendor it integratesnot measurednot published by the vendor it integrates
- Weighted mean of the five
- 74.2
- × relevance: the keyword is declared here
- 0.78
- × continuity: actively changing
- 1.00
- × evidence: widely adopted
- 1.00
- Composite score
- 57.9
Questions people ask
How do I choose between the two LinkedIn MCP servers?
Pick by scope. LinkedIn MCP Server (stickerdaniel/linkedin-mcp-server) is aimed at profiles, companies, jobs, messaging, feed and connection requests, and it had 4,119 GitHub stars, 9,464 PyPI downloads in a typical week, a tool description grade of A and 436 commits in the last 12 weeks. LinkedIn MCP Server (eliasbiondo/linkedin-mcp-server) is narrower, aimed at extracting people, company and job data from a logged-in session, and it had 212 GitHub stars, a tool description grade of A, a commit 12 days ago and 4 commits in the last 12 weeks.
Do these servers need API keys or a logged-in session?
It depends on the server, and the READMEs differ. Reddit Buddy MCP requires no API keys and treats authentication as optional for higher rate limits, and Reddit MCP Server (eliasbiondo/reddit-mcp-server) also works without API keys. Both LinkedIn servers drive browser automation from a logged-in session, and x-use drives a real Chrome session using exported cookies instead of an X API key.
Which Reddit server should I use if I only need to read?
Reddit Buddy MCP is the most established read-focused option here, with 874 GitHub stars, 1,810 npm downloads in a typical week, 5 tools, a tool description grade of A and a commit 3 days ago. Reddit MCP Server (eliasbiondo/reddit-mcp-server) is a lighter read-only choice with 166 GitHub stars, 298 PyPI downloads in a typical week, a tool description grade of A and a commit 12 days ago. Reddit MCP Server (dialog-tools/reddit-research-mcp) is aimed at multi-community research with citations and had a commit 16 days ago.
Can these servers publish, or only read?
Several publish. reddit-mcp-server (jordanburke/reddit-mcp-server) exposes create_post and reply_to_post among 12 tools, with a tool description grade of B and a commit 5 days ago. Twitter MCP Server (EnesCinr/twitter-mcp) posts tweets with up to four images and searches Twitter, with 3 tools, a tool description grade of B, 407 GitHub stars and a commit 2 days ago. x-use goes further on safety by staging writes as drafts by default, and its repository had a commit 3 days ago.
What should I be careful about with the older Facebook servers?
Check the commit record before you build on one. Facebook-Ads-MCP-Server (gomarble-ai/facebook-ads-mcp-server) is labelled Abandoned but popular and its last commit was 92 days ago, with 0 commits in the last 12 weeks, although it still has 366 GitHub stars. facebook-mcp-server (HagaiHen/facebook-mcp-server) carries the same label, with its last commit 196 days ago, 0 commits in the last 12 weeks and 234 GitHub stars. Neither repository is archived.
Is there a server for TikTok data?
TikTok MCP (Seym0n/tiktok-mcp) covers video subtitles, post details such as likes, hashtags and publishing time, and keyword search. It exposes 3 tools with a tool description grade of A, and it had 208 GitHub stars, 8 stars gained per 30 days and a commit 78 days ago, with 1 commit in the last 12 weeks. That is a much quieter commit record than the LinkedIn, Reddit and X options above, so expect a stable rather than fast-moving server.
Other comparisons
- Best Figma MCP servers
- Best Obsidian MCP servers
- Best YouTube MCP servers
- Best Jira MCP servers
- Best PostgreSQL MCP servers
- Best MySQL MCP servers
- Best GitHub MCP servers
- Best Notion MCP servers
- Best Linear MCP servers
- Best Discord MCP servers
- Best MongoDB MCP servers
- Best CRM MCP servers
- Best Payments MCP servers
- Best Ecommerce MCP servers
- Best Browser Automation MCP servers
- Best Email MCP servers
- Best PDF MCP servers
- Best Kubernetes MCP servers
- Best Docker MCP servers
- Best AWS MCP servers
- Best Spreadsheets MCP servers
- Best SEO MCP servers
- Best Web Search MCP servers
- Best Web Scraping MCP servers
- Best Vector Database MCP servers
- Best Code Execution MCP servers
- Best Calendar MCP servers
- Best Observability MCP servers
- Best Image Generation MCP servers
- Best Speech & Transcription MCP servers
- Best Maps MCP servers
- Best Travel MCP servers
- Best Stock Market MCP servers
- Best Web Analytics MCP servers
- Best Memory MCP servers
- Best Documentation MCP servers
- Best Security MCP servers