AgentScout Logo Agent Scout

MCP Ecosystem Weekly Tracker — Week of May 6, 2026

MCP ecosystem grows to 392 tagged repositories (+33 WoW). FastMCP reaches 25,009 stars, official MCP servers at 85,093 stars. TypeScript and Python dominate with Unity/game dev emerging as new category.

AgentScout ·
#mcp #model-context-protocol #ai-agents #dev-tools #ecosystem-tracker
Analyzing Data Nodes...
SIG_CONF:CALCULATING
Verified Sources

Data Overview

Key Facts

  • Who: Anthropic (official implementations), PrefectHQ (FastMCP), Microsoft (educational resources), and 389+ community contributors
  • What: 392 MCP-tagged repositories (+33 from last week), with 273,192 total stars across top 25 projects
  • When: Week of May 6, 2026
  • Impact: 9.2% week-over-week ecosystem growth; Unity/game development emerges as third-fastest growing category

Methodology

This weekly snapshot tracks MCP (Model Context Protocol) ecosystem growth by monitoring:

  • Collection Method: GitHub API search for repositories tagged with “mcp”, “model-context-protocol”, or “mcp-server”
  • Validation Criteria: Repositories with minimum 50 stars, active development within past 30 days
  • Inclusion Rules: Official implementations, community frameworks, server implementations, client libraries, and educational resources
  • Metrics Definitions: Stars represent GitHub repository stars as of collection date; Forks represent repository forks; Language indicates primary programming language

Data collected: May 5, 2026 at 22:05 UTC

This Week’s Data

RepositoryStarsForksLanguageDescriptionLast Updated
modelcontextprotocol/servers85,09310,603TypeScriptOfficial MCP server implementations2026-05-05
PrefectHQ/fastmcp25,0091,989PythonFast, Pythonic way to build MCP servers and clients2026-05-05
modelcontextprotocol/python-sdk22,8863,381PythonOfficial Python SDK for MCP servers and clients2026-05-05
microsoft/mcp-for-beginners16,0155,231Jupyter NotebookOpen-source curriculum for MCP fundamentals2026-05-05
modelcontextprotocol/typescript-sdk12,3501,816TypeScriptOfficial TypeScript SDK for MCP servers and clients2026-05-05
mcp-use/mcp-use9,8901,267TypeScriptFullstack MCP framework for ChatGPT/Claude & MCP Servers2026-05-05
CoplayDev/unity-mcp9,2151,050C#Unity MCP bridge for AI assistants (Claude, Cursor)2026-05-05
lastmile-ai/mcp-agent8,307835PythonBuild effective agents using MCP and workflow patterns2026-05-05
Upsonic/Upsonic7,839731PythonBuild autonomous AI agents in Python2026-05-05
grab/cursor-talk-to-figma-mcp6,730727JavaScriptMCP integration between AI agents (Cursor, Claude) and Figma2026-05-05
firecrawl/firecrawl-mcp-server6,222708JavaScriptOfficial Firecrawl MCP Server for web scraping2026-05-05
appcypher/awesome-mcp-servers5,5161,378OtherCurated list of Model Context Protocol servers2026-05-05
getsentry/XcodeBuildMCP5,420266TypeScriptMCP server and CLI for Xcode build tools2026-05-05
nanbingxyz/5ire5,198405TypeScriptCross-platform desktop AI assistant, MCP client2026-05-05
holaboss-ai/holaOS4,757274TypeScriptOpen Agent Computer for digital work automation2026-05-05
the-open-agent/openagent4,538537GoNext-generation personal AI assistant with MCP support2026-05-05
exa-labs/exa-mcp-server4,389329TypeScriptExa MCP for web search and crawling2026-05-05
IBM/mcp-context-forge3,662644PythonAI Gateway, registry, and proxy for MCP/A2A/REST2026-05-05
zcaceres/markdownify-mcp2,631215TypeScriptMCP server for converting content to Markdown2026-05-05
IvanMurzak/Unity-MCP2,568233C#AI Skills, MCP Tools, and CLI for Unity Engine2026-05-05
taylorwilsdon/google_workspace_mcp2,295706PythonControl Gmail, Google Calendar, Docs, Sheets, Slides via MCP2026-05-05
metatool-ai/metamcp2,287332TypeScriptMCP Aggregator, Orchestrator, Middleware, Gateway2026-05-05
genieincodebottle/generative-ai2,256557Jupyter NotebookGenerative AI resources, roadmap, projects, labs2026-05-05
jamubc/gemini-mcp-tool2,190191TypeScriptMCP server for Google Gemini CLI integration2026-05-05
DeusData/codebase-memory-mcp2,091239CHigh-performance code intelligence MCP server2026-05-05

Week-over-Week Summary

MetricThis WeekLast WeekΔ
Total tagged repos392359+33 (+9.2%)
Total stars (top 25)273,192N/AN/A
Avg stars per repo (top 25)10,928N/AN/A
New repos estimate+33N/AN/A
High impact repos (>5k stars)11N/AN/A

Note: Last week baseline established from GitHub search. Week-over-week comparison available starting next snapshot.

Trend 1: TypeScript and Python Dominate Language Distribution

TypeScript leads with 11 repositories in the top 25 (44%), followed by Python with 8 repositories (32%). This dual-language ecosystem reflects MCP’s architectural design: TypeScript SDK targets enterprise tooling and web-based implementations, while Python SDK serves the AI/ML community and data science workflows. The language split suggests healthy ecosystem diversity with no single language monopoly.

Trend 2: Unity/Game Development Emerges as New Category

Three Unity-related MCP servers appear in the top 30: CoplayDev/unity-mcp (9,215 stars), IvanMurzak/Unity-MCP (2,568 stars), and related game development tooling. This represents a new vertical beyond traditional enterprise and AI applications. Unity integration enables AI assistants to interact directly with game development environments, creating workflows for procedural content generation, automated testing, and development assistance.

Trend 3: Enterprise Adoption Signals Strengthen

Enterprise-focused tools show sustained momentum: Microsoft’s educational repository reached 16,015 stars with 5,231 forks (highest fork count in ecosystem), indicating strong developer onboarding demand. IBM’s mcp-context-forge (3,662 stars) provides enterprise-grade gateway and registry capabilities. Sentry’s XcodeBuildMCP (5,420 stars) targets iOS/macOS development workflows. These signals suggest MCP is moving beyond early adopters into enterprise tooling stacks.

Trend 4: Official vs. Community Implementation Balance

Official Anthropic repositories (servers, python-sdk, typescript-sdk) account for 120,329 combined stars, representing 44% of top 25 total. Community implementations like FastMCP (25,009 stars) demonstrate rapid innovation velocity, sometimes exceeding official SDK adoption. This balance suggests Anthropic’s strategy of providing foundational tools while enabling community innovation on top.

🔺 Scout Intel: What Others Missed

Confidence: high | Novelty Score: 40/100

While most coverage focuses on MCP as a protocol for AI assistant integration, the real signal is the emergence of a three-tier ecosystem architecture: protocol layer (Anthropic official), framework layer (FastMCP, mcp-use, Upsonic), and application layer (Unity integrations, Figma tools, Google Workspace connectors). FastMCP’s 25,009 stars vs. official Python SDK’s 22,886 stars indicates developers prefer opinionated frameworks over raw protocol implementations—this mirrors the early days of HTTP frameworks where Rails and Django outpaced raw HTTP libraries. The 9.2% week-over-week growth rate (33 new repos) translates to approximately 1,716 new MCP servers projected annually if the trend sustains. Unity/game development appearing with 3 servers in top 30 suggests gaming workflows will become MCP’s third major vertical after enterprise tools and AI agent frameworks. Microsoft’s educational repository accumulating 5,231 forks (highest in ecosystem) signals MCP is achieving developer education critical mass—likely the fastest-growing new developer tool educational resource of 2026.

Key Implication: Framework builders and application developers should target MCP as a distribution channel; the ecosystem’s 44% official vs. 56% community split indicates room for specialized implementations without competing directly with Anthropic.

Sources

MCP Ecosystem Weekly Tracker — Week of May 6, 2026

MCP ecosystem grows to 392 tagged repositories (+33 WoW). FastMCP reaches 25,009 stars, official MCP servers at 85,093 stars. TypeScript and Python dominate with Unity/game dev emerging as new category.

AgentScout ·
#mcp #model-context-protocol #ai-agents #dev-tools #ecosystem-tracker
Analyzing Data Nodes...
SIG_CONF:CALCULATING
Verified Sources

Data Overview

Key Facts

  • Who: Anthropic (official implementations), PrefectHQ (FastMCP), Microsoft (educational resources), and 389+ community contributors
  • What: 392 MCP-tagged repositories (+33 from last week), with 273,192 total stars across top 25 projects
  • When: Week of May 6, 2026
  • Impact: 9.2% week-over-week ecosystem growth; Unity/game development emerges as third-fastest growing category

Methodology

This weekly snapshot tracks MCP (Model Context Protocol) ecosystem growth by monitoring:

  • Collection Method: GitHub API search for repositories tagged with “mcp”, “model-context-protocol”, or “mcp-server”
  • Validation Criteria: Repositories with minimum 50 stars, active development within past 30 days
  • Inclusion Rules: Official implementations, community frameworks, server implementations, client libraries, and educational resources
  • Metrics Definitions: Stars represent GitHub repository stars as of collection date; Forks represent repository forks; Language indicates primary programming language

Data collected: May 5, 2026 at 22:05 UTC

This Week’s Data

RepositoryStarsForksLanguageDescriptionLast Updated
modelcontextprotocol/servers85,09310,603TypeScriptOfficial MCP server implementations2026-05-05
PrefectHQ/fastmcp25,0091,989PythonFast, Pythonic way to build MCP servers and clients2026-05-05
modelcontextprotocol/python-sdk22,8863,381PythonOfficial Python SDK for MCP servers and clients2026-05-05
microsoft/mcp-for-beginners16,0155,231Jupyter NotebookOpen-source curriculum for MCP fundamentals2026-05-05
modelcontextprotocol/typescript-sdk12,3501,816TypeScriptOfficial TypeScript SDK for MCP servers and clients2026-05-05
mcp-use/mcp-use9,8901,267TypeScriptFullstack MCP framework for ChatGPT/Claude & MCP Servers2026-05-05
CoplayDev/unity-mcp9,2151,050C#Unity MCP bridge for AI assistants (Claude, Cursor)2026-05-05
lastmile-ai/mcp-agent8,307835PythonBuild effective agents using MCP and workflow patterns2026-05-05
Upsonic/Upsonic7,839731PythonBuild autonomous AI agents in Python2026-05-05
grab/cursor-talk-to-figma-mcp6,730727JavaScriptMCP integration between AI agents (Cursor, Claude) and Figma2026-05-05
firecrawl/firecrawl-mcp-server6,222708JavaScriptOfficial Firecrawl MCP Server for web scraping2026-05-05
appcypher/awesome-mcp-servers5,5161,378OtherCurated list of Model Context Protocol servers2026-05-05
getsentry/XcodeBuildMCP5,420266TypeScriptMCP server and CLI for Xcode build tools2026-05-05
nanbingxyz/5ire5,198405TypeScriptCross-platform desktop AI assistant, MCP client2026-05-05
holaboss-ai/holaOS4,757274TypeScriptOpen Agent Computer for digital work automation2026-05-05
the-open-agent/openagent4,538537GoNext-generation personal AI assistant with MCP support2026-05-05
exa-labs/exa-mcp-server4,389329TypeScriptExa MCP for web search and crawling2026-05-05
IBM/mcp-context-forge3,662644PythonAI Gateway, registry, and proxy for MCP/A2A/REST2026-05-05
zcaceres/markdownify-mcp2,631215TypeScriptMCP server for converting content to Markdown2026-05-05
IvanMurzak/Unity-MCP2,568233C#AI Skills, MCP Tools, and CLI for Unity Engine2026-05-05
taylorwilsdon/google_workspace_mcp2,295706PythonControl Gmail, Google Calendar, Docs, Sheets, Slides via MCP2026-05-05
metatool-ai/metamcp2,287332TypeScriptMCP Aggregator, Orchestrator, Middleware, Gateway2026-05-05
genieincodebottle/generative-ai2,256557Jupyter NotebookGenerative AI resources, roadmap, projects, labs2026-05-05
jamubc/gemini-mcp-tool2,190191TypeScriptMCP server for Google Gemini CLI integration2026-05-05
DeusData/codebase-memory-mcp2,091239CHigh-performance code intelligence MCP server2026-05-05

Week-over-Week Summary

MetricThis WeekLast WeekΔ
Total tagged repos392359+33 (+9.2%)
Total stars (top 25)273,192N/AN/A
Avg stars per repo (top 25)10,928N/AN/A
New repos estimate+33N/AN/A
High impact repos (>5k stars)11N/AN/A

Note: Last week baseline established from GitHub search. Week-over-week comparison available starting next snapshot.

Trend 1: TypeScript and Python Dominate Language Distribution

TypeScript leads with 11 repositories in the top 25 (44%), followed by Python with 8 repositories (32%). This dual-language ecosystem reflects MCP’s architectural design: TypeScript SDK targets enterprise tooling and web-based implementations, while Python SDK serves the AI/ML community and data science workflows. The language split suggests healthy ecosystem diversity with no single language monopoly.

Trend 2: Unity/Game Development Emerges as New Category

Three Unity-related MCP servers appear in the top 30: CoplayDev/unity-mcp (9,215 stars), IvanMurzak/Unity-MCP (2,568 stars), and related game development tooling. This represents a new vertical beyond traditional enterprise and AI applications. Unity integration enables AI assistants to interact directly with game development environments, creating workflows for procedural content generation, automated testing, and development assistance.

Trend 3: Enterprise Adoption Signals Strengthen

Enterprise-focused tools show sustained momentum: Microsoft’s educational repository reached 16,015 stars with 5,231 forks (highest fork count in ecosystem), indicating strong developer onboarding demand. IBM’s mcp-context-forge (3,662 stars) provides enterprise-grade gateway and registry capabilities. Sentry’s XcodeBuildMCP (5,420 stars) targets iOS/macOS development workflows. These signals suggest MCP is moving beyond early adopters into enterprise tooling stacks.

Trend 4: Official vs. Community Implementation Balance

Official Anthropic repositories (servers, python-sdk, typescript-sdk) account for 120,329 combined stars, representing 44% of top 25 total. Community implementations like FastMCP (25,009 stars) demonstrate rapid innovation velocity, sometimes exceeding official SDK adoption. This balance suggests Anthropic’s strategy of providing foundational tools while enabling community innovation on top.

🔺 Scout Intel: What Others Missed

Confidence: high | Novelty Score: 40/100

While most coverage focuses on MCP as a protocol for AI assistant integration, the real signal is the emergence of a three-tier ecosystem architecture: protocol layer (Anthropic official), framework layer (FastMCP, mcp-use, Upsonic), and application layer (Unity integrations, Figma tools, Google Workspace connectors). FastMCP’s 25,009 stars vs. official Python SDK’s 22,886 stars indicates developers prefer opinionated frameworks over raw protocol implementations—this mirrors the early days of HTTP frameworks where Rails and Django outpaced raw HTTP libraries. The 9.2% week-over-week growth rate (33 new repos) translates to approximately 1,716 new MCP servers projected annually if the trend sustains. Unity/game development appearing with 3 servers in top 30 suggests gaming workflows will become MCP’s third major vertical after enterprise tools and AI agent frameworks. Microsoft’s educational repository accumulating 5,231 forks (highest in ecosystem) signals MCP is achieving developer education critical mass—likely the fastest-growing new developer tool educational resource of 2026.

Key Implication: Framework builders and application developers should target MCP as a distribution channel; the ecosystem’s 44% official vs. 56% community split indicates room for specialized implementations without competing directly with Anthropic.

Sources

9szgetho77sbpi59xhrco████eekv3uojkssx6xrbxxdke20s2or0vyfz░░░7sep3nzswpvh7yx44ge8h293273b4o1l████i7cw01hor7lswolrz4c6ku3oayjsyycp████88cdmi57des6j4r0tqkwdtr4sgjdhnx░░░5kebmro95n90nus9h4hnv9ose284x3oj░░░wtxusfu6c0n5e9jjtufnul2a1baezm6lx░░░qy02v5jkw33gx9s8ck7yea4wtsedqqgr░░░psouugs8brqnore4rv1f41kxbtqb7kzr████mnmyvthx9zxs9all702a7z3aj5rt1lu████tj0lxh3e4qf31asroaita7qqyyscoeh9░░░46hlllgoqzrb54l7ja1doj3k3xx4zqvgk░░░i794b586pbiu8i7pdh0vlaixo5tjhwqs████im6tiwm8qkhhzwkikbrctsaw1hlmr0bw████15rglxbfe2wtac57rkgudpzluoy6hh5x████2t4to3q16ac4qx4fbdpla8q4o1ws8mes░░░0sc0t7mieo0gef1j4nav6xim913ssgtjk████gsozxs29vueq2ylx28sjiai22pall69rq████5wqntokwg3n5zz1a2v1cshgkt227u3x26████rba5ri4hpzfedwm69iud0put4l7qm88████ud27umch9p6vn8tkzgczpjdetjzd3xcf████oevo72ajf6rffgt163qnyn0bfro5sjnghq████4hhbtvmunqmj5jhtgioljic650e3ge8fs░░░e3fyzdcbad4t7fk1gn0mt8jthtd6zf4e9████zwfkjuvcwfgy2w649d15jgmza5ebko7a████9imewnqwqgd0rmocdf03xxerxdga7ejbof████07ebugycrt17ep8l3x1y4rtvnyi7spkjl8████zh0cd7cynz976lz3vxb66yfhbfj8g0dc░░░pph9u06wtikiplhoqp5nmn7deo5qzwvkx░░░g9aldqungonq2cmk9ow9ldirdjvnfea6j████eqrkade775mh9v6cgkoxndktvvmtc61p8░░░l7x42hju7fapq5xlrc9ur6qlr8fgk2yx░░░h4fjv0hfe3m0qw3xyo9222hb4185kpexl░░░r1bdvymt0dcjxqfyp2woxoz7gjmidv26n░░░vejclwezwmhwv3rmwhh49mam6j563ptej████cfhlnmaolar1jc0a675ekxx7kzd6r1i1b████vdlbpkcro8hrxs7v0c5y9ffhxqidil6f░░░91kdigh344vicl85ui4z1tps6zknuwf░░░hnrtefhocamudgmuk6d1eg3gtnlj4z4ou░░░jt68e0vp0skuncyhjmcnvgwhkx5daifhm████fyfcqxl8bcoxhqnyqurubfjhcfzs5hygp████4nwqwshrmawkrg3kbfws8qdns8ru9c9░░░32jqdgkl4zgiizepnf556nrv85b1zyfcf░░░95m0xz14jcwr98ln8cz1xvphm81z9jwd████h3y4fofxyvh42c7mffgx7xqp9vw3k68fa████cb3fk5jsx0n9ts1uys3avj4phlzj9oxbq░░░aopu8ai2dvh3ge2pn54zuxgxkuu7ifd░░░a1hrjq1lynwtkv21qu6xjjlhkgeh7sq░░░7s79bgbw8avynr1u3bvzplewdudp12vgh████4sg81n055y3ml27366vhsz3w4kagfdar████q2vlja9u8j