Context7 MCP. Ground your AI agent in real-time API docs.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Context7. Gives your AI agent access to up-to-date, version-specific technical documentation and code examples for any library or framework. Stop relying on outdated LLM training data; use Context7 to pull raw Markdown docs and valid code snippets directly into your development flow, whether you're using React, Next.js, or Tailwind CSS.
What your AI agents can do
Query docs
Retrieves documentation and code examples for a specific library ID about a given topic.
Resolve library
Finds the correct, exact library ID and latest version matching a given framework or library search query.
It takes a general framework name and outputs a precise, versioned path needed to fetch accurate documentation.
It analyzes a given library ID and retrieves raw Markdown documentation chunks about a specific topic.
It pulls valid, ready-to-use code snippets for components or functions matching the requested documentation.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Context7 MCP Server: 2 Tools for Dev Docs
Use these tools to resolve framework versions and retrieve documentation chunks or code examples for specific libraries.
019d757bquery docs
Retrieves documentation and code examples for a specific library ID about a given topic.
019d757bresolve library
Finds the correct, exact library ID and latest version matching a given framework or library search query.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Context7, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
Your AI client needs the latest API specs and code snippets, not stale LLM knowledge. Context7 gives your agent direct access to current, version-specific documentation and working code for any framework. You can stop guessing at API calls; your agent pulls raw Markdown and valid code right into your dev flow, whether you're messing with React, Next.js, or Tailwind CSS.
To get started, you'll use two core tools. First, resolve_library takes a general framework name and spits out the exact, versioned path needed to pull accurate docs. Then, query_docs takes that precise library ID and digs up raw Markdown documentation chunks about whatever topic you need. You can also use query_docs to pull ready-to-use code snippets for functions or components.
When your agent runs resolve_library with a framework name, it outputs a precise path, like /facebook/react/18.2.0. You then feed that path into query_docs. This lets your agent pull raw docs on a specific topic, and it also lets it pull valid code examples for components or functions matching that topic.
This means your agent never relies on outdated training data; it works off the current truth of the library.
If you're working with a component, you can first use resolve_library to nail down the version, then use query_docs to get the documentation for a specific variable, and finally use query_docs again to extract a working code example for that component. It keeps your whole development cycle grounded in reality.
You're always working with the exact version of the library you're targeting. This saves you from hours of debugging because your agent knows the right syntax and the latest API signatures. You're getting a live, verifiable source of truth for your code, not just a guess.
Context7 works as a documentation-specialized RAG layer. Your agent pulls the technical data you need, making sure everything it writes is accurate to the library's current state. You just point your AI client at the server, and it handles the rest. You get the real deal, every time.
How Context7 MCP Works
- 1 First, use the
resolve_librarytool. You give it a framework name (e.g., 'nextjs'). It returns a list of specific, versioned IDs (e.g., 'vercel/next.js/14.1.0'). - 2 Next, take one of those specific IDs and the topic you need. You pass both to the
query_docstool. - 3 The tool returns raw Markdown documentation and executable code examples for that exact version and topic.
The bottom line is: You use resolve_library to nail down the exact version, then query_docs to get the accurate context.
Who Is Context7 MCP For?
Software developers, AI engineers, and technical writers need this. If your job involves building software that relies on specific, changing APIs, you're the target. Context7 handles the biggest headache: keeping your knowledge base current and accurate without manual updates.
Needs to pull specific API documentation and code snippets without leaving the IDE or searching through multiple online docs.
Requires grounding coding agents in real-time technical documentation to make sure the generated code actually works with current API signatures.
Needs to verify version-specific API signatures and cross-reference documentation blocks quickly before publishing a guide.
What Changes When You Connect
- Stop relying on LLMs' internal knowledge cutoffs. Context7 pulls documentation synchronized with the latest releases, ensuring your agent uses current API signatures.
- Pinpoint exact library versions. The
resolve_librarytool resolves vague names like 'react' into deterministic paths (e.g., /facebook/react/18.2.0), eliminating guesswork. - Get ready-to-use code. The
query_docstool pulls valid, version-specific code examples directly into your development flow, saving copy-pasting time. - Build accurate coding agents. Use Context7 as a documentation-specialized RAG layer. Your agent won't hallucinate outdated functions.
- Work without context switching. Developers pull API documentation and code snippets directly into the IDE, avoiding manual searches.
- Cross-reference technical facts. Technical writers can use this to verify version-specific API signatures and quickly compare documentation blocks.
Real-World Use Cases
The developer needs the latest Next.js setup.
A developer needs to implement the 'App Router' feature. They ask their agent, and the agent first uses resolve_library to get the Next.js 14.1.0 ID. Then, it runs query_docs to retrieve official Markdown segments detailing the required layout.js and page.js structure with working code examples.
The AI agent needs to use Tailwind CSS v4.
An AI engineer is writing a component and needs to use a new feature in Tailwind CSS v4. They prompt the agent, which uses Context7 to query v4.0.0 documentation. The agent then provides the exact code snippets for the new high-performance engine, ensuring the code is current.
The technical writer is verifying an API.
A technical writer must confirm the signature for a function in a major library. Instead of searching multiple online pages, they use Context7. The agent runs query_docs to pull the raw, version-specific documentation, guaranteeing the API signature is correct for the target library version.
The rapid prototypers need boilerplate code.
A rapid developer is starting a new microservice using a new framework. They ask the agent to pull component examples. The agent uses Context7 to retrieve instant boilerplates and component examples, allowing them to build the structure without manual setup.
The Tradeoffs
Relying on general LLM knowledge
The agent hallucinates an API signature for React because the LLM was trained on old documentation, leading to a compilation error when the developer runs the code.
→
Don't trust general knowledge. First, use resolve_library to nail down the exact version (e.g., 'vercel/next.js/14.1.0'). Then, use query_docs with that specific ID to get the truth.
Searching documentation manually
The developer opens 10 browser tabs, copies snippets, and manually compares versions across different online guides, wasting hours.
→ Keep the search inside your IDE. Use the Context7 MCP Server to funnel the documentation directly to your agent. It handles the versioning and retrieval automatically.
Confusing general search with code context
The developer asks the agent, 'How do I use X?' without specifying a library. The agent gives general conceptual advice that isn't executable code.
→
Always start by running resolve_library to confirm the library and version. Then, pass the specific ID to query_docs to get an actionable, contextual answer.
When It Fits, When It Doesn't
Use this if your work requires code to be correct to the minute. You need a guaranteed source of truth for APIs, frameworks, or libraries that update frequently. If you are building a complex system and can't afford a build failure because of an outdated API call, this tool is critical. Don't use it if you just need general conceptual advice or high-level architectural patterns; those don't require specific versions. If you just need to search general concepts, an existing search engine is fine. But if you need code that works right now, use Context7.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Context7. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 2 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Finding the right API documentation shouldn't require a dozen browser tabs.
Before Context7, finding specific technical documentation was a nightmare. You’d start with a general search, click through a dozen links, and manually compare versions across different sites. You'd spend time copy-pasting snippets and cross-referencing API signatures just to confirm if a function existed in the version you were actually using.
Now, you connect your agent to Context7. You tell it the framework and the concept. Context7 resolves the exact version and pulls the raw Markdown and code examples right into your IDE. You get actionable, verified context without leaving your workflow.
Context7 MCP Server: Get verifiable code context.
The two manual steps that disappear are: 1) Determining the correct, versioned library path, and 2) Sifting through outdated docs to find a working example. You never have to guess which version you're using again.
It’s immediate. You ask for the documentation, and it returns the raw, version-locked context. You don't just get information; you get code that works with the latest releases.
Common Questions About Context7 MCP
How does Context7 MCP Server handle different library versions? +
Context7 resolves the library version first. You use resolve_library to pinpoint the exact version (e.g., 14.1.0). Then, query_docs fetches documentation locked to that specific version, preventing version mismatch errors.
Can I use Context7 MCP Server to find code for a framework I don't know? +
Yes. You start by using resolve_library. It searches for the framework name and returns all matching, deterministic IDs, helping you select the correct starting point.
Is the documentation from Context7 MCP Server up-to-date? +
Yes. Context7 is designed to pull documentation synchronized with the latest releases, bypassing the knowledge cutoffs of standard LLMs.
What if my library name is ambiguous? +
If the name is ambiguous, resolve_library gives you a list of deterministic paths and versions (e.g., v12, v13, v14). You then select the version you need for query_docs.
Do I need to manually enter the API key for Context7 MCP Server? +
Yes. You must subscribe to the server and input your Context7 API Key, which you find in your Context7/Upstash Dashboard.
How do I use the `resolve_library` tool in Context7 to find a library's exact path? +
The resolve_library tool finds the deterministic path and version ID needed for deep documentation fetching. Just provide the framework name (like 'react' or 'nextjs') and the tool returns all matching IDs and versions, letting you pick the right one.
What happens if I query documentation with an invalid library ID using Context7? +
Context7 validates the ID and returns a clear error message, telling you exactly what went wrong. This prevents your agent from trying to read non-existent documentation and keeps your workflow clean.
Is Context7 designed to handle complex or multi-part technical queries? +
Yes, Context7 handles complex queries by first using resolve_library to pin down the exact context. Then, the query_docs tool pulls specific Markdown chunks—like showing how to use a feature—so your agent gets precise, actionable data.
Can my agent find the latest documentation for a specific Tailwind CSS version? +
Yes. First, use the 'resolve_library' tool with 'tailwindcss'. It will return the deterministic ID and version (e.g., 'tailwindcss/3.4.1'). Then, use 'query_docs' to pull the exact Markdown blocks for your specific topic.
Does Context7 provide code examples for the libraries I search for? +
Absolutely. When you query documentation via the 'query_docs' tool, the agent retrieves not only textual descriptions but also version-specific code examples found in the original library documentation to ensure implementation accuracy.
How does this help prevent AI hallucinations in coding tasks? +
Standard LLMs have a training data cutoff. Context7 pulls live, version-specific documentation chunks that act as ground-truth context. By grounding your agent in this real-time data, it avoids hallucinating outdated or non-existent API methods.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Pipedream
Manage Pipedream serverless workflows, sources, webhooks, and raw event data natively via AI agents.
Kong Gateway
Manage your API Gateway infrastructure — list services, configure routes, and manage consumers or plugins directly from any AI agent.
Kong (AI API Gateway)
Manage your API Gateway via Kong — orchestrate services, routes, and AI plugins directly from your agent.
You might also like
Radar
Equip your AI with geolocation tools — convert coordinates to addresses, map intellectual territories, calculate routes, and perform IP-based lookups.
LangSmith (LLM Observability & Hub)
Monitor LLM apps via LangSmith — track traces, audit prompt templates, and manage evaluation datasets.
Gupshup Messaging
Send WhatsApp messages, templates, media, and reactions via Gupshup API.