How do I use AI Context MCP?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@AI Context MCP show me the streaming API for OpenAI Python v1.0.0" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

AI Context MCP

by Prathmeshkunturwar

Overview Schema Source Related Servers Discussions Admin

TypeScript

Hybrid

Fetches up-to-date documentation, changelogs, and source signatures for the CrewAI agent framework.

Fetches up-to-date documentation, changelogs, and source signatures for the Drizzle ORM.

Fetches up-to-date documentation, changelogs, and source signatures for Express.js.

Fetches up-to-date documentation, changelogs, and source signatures for Fastify.

Fetches up-to-date documentation, changelogs, and source signatures for Firebase JS SDK.

Fetches up-to-date documentation, changelogs, and source signatures for Hono.

Fetches up-to-date documentation, changelogs, and source signatures for LangChain (Python and JavaScript).

Fetches up-to-date documentation, changelogs, and source signatures for LangGraph.

Fetches up-to-date documentation, changelogs, and source signatures for Mongoose ODM.

Fetches up-to-date documentation, changelogs, and source signatures for Next.js.

Fetches up-to-date documentation, changelogs, and source signatures for Nuxt.

Fetches up-to-date documentation, changelogs, and source signatures for Ollama.

Fetches up-to-date documentation, changelogs, and source signatures for OpenAI SDKs (Python and Node.js).

Fetches up-to-date documentation, changelogs, and source signatures for Prisma ORM.

Fetches up-to-date documentation, changelogs, and source signatures for Pydantic and Pydantic AI.

Fetches up-to-date documentation, changelogs, and source signatures for PyTorch.

Fetches up-to-date documentation, changelogs, and source signatures for React.

Fetches up-to-date documentation, changelogs, and source signatures for Supabase JS SDK.

Fetches up-to-date documentation, changelogs, and source signatures for Svelte and SvelteKit.

Fetches up-to-date documentation, changelogs, and source signatures for tRPC.

Fetches up-to-date documentation, changelogs, and source signatures for TypeORM.

Fetches up-to-date documentation, changelogs, and source signatures for the Vercel AI SDK.

Fetches up-to-date documentation, changelogs, and source signatures for Vitest.

Fetches up-to-date documentation, changelogs, and source signatures for Zod.

Related MCP server: Context7 MCP

Why AI Context MCP?

LLMs are trained on static snapshots of the web. By the time you use Claude or Cursor, its knowledge of langchain, openai, pytorch etc. is already months out of date — causing:

❌ Hallucinated function signatures that don't exist
❌ Deprecated patterns the model confidently recommends
❌ Silent breaking changes that crash your app at runtime

AI Context MCP solves this by fetching current docs and changelogs from GitHub at query time, ranking them semantically, and injecting them into the model's context window.

Quick Start

Option 1 — NPM Global Install (recommended)

npm install -g ai-context-mcp

Option 2 — Clone & Build

git clone https://github.com/Prathmeshkunturwar/Context_Mcp.git
cd Context_Mcp
npm install && npm run build

Setup

Claude Code

Add to your project's .mcp.json:

{
  "mcpServers": {
    "ai-context": {
      "command": "npx",
      "args": ["-y", "ai-context-mcp"],
      "env": { "GITHUB_TOKEN": "ghp_your_token_here" }
    }
  }
}

Claude Desktop

Edit claude_desktop_config.json:

OS	Path
Windows	`%APPDATA%\Claude\claude_desktop_config.json`
macOS	`~/Library/Application Support/Claude/claude_desktop_config.json`
Linux	`~/.config/Claude/claude_desktop_config.json`

{
  "mcpServers": {
    "ai-context": {
      "command": "npx",
      "args": ["-y", "ai-context-mcp"],
      "env": { "GITHUB_TOKEN": "ghp_your_token_here" }
    }
  }
}

Cursor / Other MCP Clients

{
  "mcpServers": {
    "ai-context": {
      "command": "npx",
      "args": ["-y", "ai-context-mcp"],
      "env": { "GITHUB_TOKEN": "ghp_your_token_here" }
    }
  }
}

Environment Variables

Variable	Required	Description
`GITHUB_TOKEN`	Recommended	GitHub PAT. Without it: 60 req/hr. With it: 5,000/hr. Create one (no scopes needed).
`TRANSPORT`	Optional	Set to `http` for Express HTTP server instead of stdio.
`CACHE_TTL_HOURS`	Optional	Cache TTL in hours (default: 24).
`PORT`	Optional	HTTP port when `TRANSPORT=http` (default: 3000).

Tools Reference

`resolve-library-id`

Maps natural language to a registry library ID.

"LangChain JavaScript" → /langchain-ai/langchainjs

`query-docs`

Fetches version-pinned documentation snippets, semantically ranked using MiniLM-L6.

query-docs(libraryId: "/openai/openai-python", query: "streaming", version: "1.0.0")

`get-changelog-diff`

Extracts breaking changes and new features between two versions.

get-changelog-diff(libraryId: "/langchain-ai/langchainjs", fromVersion: "0.1.0", toVersion: "0.3.0")
→ 7 BREAKING CHANGES, 12 NEW FEATURES

`get-source-signature`

Extracts raw type/function/class signatures directly from source code.

get-source-signature(libraryId: "/langchain-ai/langchainjs", filePath: "...", entityName: "RunnableSequence")

`detect-project-versions`

Scans your local package.json or requirements.txt to detect installed library versions.

detect-project-versions(projectPath: "./")
→ { "langchain": "0.1.0", "@langchain/openai": "0.0.14" }

`auto-migrate-codebase`

Analyzes your codebase against changelog breaking changes and generates migration alerts.

auto-migrate-codebase(libraryId: "...", projectPath: "./", fromVersion: "0.1.0", toVersion: "0.3.0")

`suggest-skills`

Scans your project imports and recommends which libraries to look up.

suggest-skills(projectPath: "./") → ["/langchain-ai/langchainjs", "/openai/openai-node"]

Supported Libraries

🤖 LLM Provider SDKs

OpenAI Python · OpenAI Node.js · Anthropic Python · Anthropic TypeScript · Google Generative AI · Mistral AI · Cohere

🦾 Agent Frameworks

LangChain Python · LangChain.js · LangGraph · LlamaIndex · CrewAI · Pydantic AI · AutoGen · Letta (MemGPT) · Vercel AI SDK

🧠 ML Frameworks

PyTorch · HuggingFace Transformers · Diffusers · Accelerate · PEFT · Pydantic

⚡ Inference Engines

vLLM · llama.cpp · Ollama

🗄️ Vector Databases

ChromaDB · Qdrant · Pinecone · Weaviate

🌐 Web Frameworks

React · Next.js · Vue · Nuxt · Svelte · SvelteKit · Express · Fastify · Hono

🗃️ Database & ORM

Prisma · Drizzle · TypeORM · Mongoose

🔌 API & Validation

tRPC · Zod · Apollo Server

🧪 Testing & Infrastructure

Vitest · Playwright · Supabase JS · Firebase JS

57 libraries and growing! Run npm run add-library or open a PR to add yours.

Architecture

┌─────────────────────────────────────────────────────────┐
│                MCP Client (Claude / Cursor)              │
│        resolve → query → changelog → signature           │
└───────────────────────┬─────────────────────────────────┘
                        │ MCP Protocol (stdio / http)
┌───────────────────────▼─────────────────────────────────┐
│               AI Context MCP Server                      │
│                                                          │
│  ┌────────────┐  ┌────────────┐  ┌──────────────────┐   │
│  │  Registry  │  │  Fetcher   │  │ Semantic Ranker   │   │
│  │ (57 libs)  │  │ (GitHub +  │  │  (MiniLM-L6-v2)  │   │
│  └────────────┘  │  Cache)    │  └──────────────────┘   │
│                  └────────────┘                          │
│  ┌────────────┐  ┌────────────┐  ┌──────────────────┐   │
│  │ Changelog  │  │  Snippet   │  │  Source          │   │
│  │  Parser    │  │ Extractor  │  │  Signature       │   │
│  └────────────┘  └────────────┘  └──────────────────┘   │
│                                                          │
│      L1: In-Memory LRU  │  L2: SQLite (offline)         │
└───────────────────────────────────────────────────────────┘

Contributing

See CONTRIBUTING.md for full guidelines.

npm run dev          # ts-node with hot reload
npm test             # 39 Jest tests
npm run build        # Compile TypeScript
npm run add-library  # Interactive library adder

Project structure:

src/
├── server.ts              # Entry point
├── registry/libraries.json # Add libraries here
├── sources/               # GitHub, NPM, PyPI, HuggingFace
├── ranking/               # Semantic ranker + snippet extractor
├── cache/                 # Two-tier LRU + SQLite cache
└── tools/                 # changelog, versions, migrate, suggest

Troubleshooting

"Cannot find module" on start

npm run build

GitHub 403 / rate limit errors Add GITHUB_TOKEN to your env — create a token with no extra scopes.

Stale/outdated docs returned

rm -f .cache/ai-context-cache.db

First npx run hangs The MiniLM model downloads (~25MB) on first run. This is one-time only.

More help in docs/TROUBLESHOOTING.md · Open an issue

License

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityInactive

ResponsivenessNo issues

Resources

Need Help?

Related Servers

Related MCP Connectors

docfork-mcpOAuth
ai.smithery
@latest documentation and code examples to 9000+ libraries for LLMs and AI code editors in a singl…
Documentation Access Developer Tools Code Analysis
2488
llmref A
org.llmref
Versioned documentation registry and semantic search for AI tools and coding assistants.
Documentation Access Search Developer Tools
42
ref-tools-ref-tools-mcpOAuth
ai.smithery
Provide your AI coding tools with token-efficient access to up-to-date technical documentation for…
Documentation Access Developer Tools Code Analysis
21,173
Reqistry A
dev.reqistry
Discover, compare, and monitor 1,400+ APIs directly from your AI coding agent.
Developer Tools Search Monitoring
131

Related MCP Servers

Context7 MCP
Documentation Access Developer Tools RAG Systems
geobio
A
license
Not graded
quality
D
maintenance
Provides LLMs with up-to-date, version-specific documentation and code examples from library sources directly into prompts, eliminating outdated code generation and hallucinated APIs.
Updated a year ago (2025-05-29 21:17 UTC)
1,132,952 npm
MIT
Context7 MCP
Documentation Access Developer Tools RAG Systems
renCosta2025
A
license
A
quality
D
maintenance
Provides up-to-date, version-specific documentation and code examples for libraries and frameworks directly into AI prompts, eliminating outdated code generation and hallucinated APIs.
Updated a year ago (2025-09-28 19:15 UTC)
2
1,132,952 npm
MIT
Context7 MCP
Documentation Access Developer Tools RAG Systems
3737123Dush
A
license
A
quality
C
maintenance
Provides LLMs with up-to-date, version-specific documentation and code examples directly from library sources, eliminating outdated training data and hallucinated APIs by fetching current documentation at prompt time.
Updated a year ago (2025-07-08 07:19 UTC)
2
1,132,952 npm
MIT
RTFD (Read The F*****g Docs)
Documentation Access Developer Tools Search
aserper
A
license
A
quality
C
maintenance
Provides LLMs with real-time access to up-to-date documentation from PyPI, npm, crates.io, GoDocs, DockerHub, GitHub, and GCP, preventing outdated code generation and API hallucinations.
Updated 2 months ago (2026-07-16 20:17 UTC)
27
70 PyPI
14
MIT

Table of Contents

Why AI Context MCP?

Quick Start

Setup

Claude Code

Claude Desktop

Cursor / Other MCP Clients

Environment Variables

Tools Reference

resolve-library-id

query-docs

get-changelog-diff

get-source-signature

detect-project-versions

auto-migrate-codebase

suggest-skills

Supported Libraries

🤖 LLM Provider SDKs

🦾 Agent Frameworks

🧠 ML Frameworks

⚡ Inference Engines

🗄️ Vector Databases

🌐 Web Frameworks

🗃️ Database & ORM

🔌 API & Validation

🧪 Testing & Infrastructure

Architecture

Contributing

Troubleshooting

License

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

Context7 MCP

Context7 MCP

Context7 MCP

RTFD (Read The F*****g Docs)

`resolve-library-id`

`query-docs`

`get-changelog-diff`

`get-source-signature`

`detect-project-versions`

`auto-migrate-codebase`

`suggest-skills`