SEARCH & RESEARCH · FOR AI AGENTS
Context7
Retrieve library documentation and examples for an agent’s coding task.
Context7 supplies documentation search and context through MCP, REST, SDK and CLI access. Coding agents can retrieve source-linked examples and narrow requests to a library or version instead of relying only on training data.
Your integration checklist
- Capabilities
- What you need from your owner
- Connect an MCP client with OAuth or create a Context7 API key. Select allowed sources and decide whether private-repository indexing or paid overages are permitted.
- Authentication
- Bearer API key for production API requests; OAuth is supported by MCP clients. Anonymous Search API requests are only for testing and have IP-based limits.
- Your first task
- Look up one API used by the project, pass the installed library version where supported, and verify the returned source example before changing code.
- Billing & payment
- Free accounts include a monthly call allowance. Paid seats add calls and bill additional usage; private-repository parsing is charged separately. Paid-plan requests can continue into overage.
- Agent payment support
- Not verified. Programmatic usage does not establish autonomous checkout support.
- Operational limits
- Coverage depends on indexed libraries and versions. Retrieved examples do not prove compatibility with the project or guarantee correct generated code. Private sources require appropriate access.
ORIGINAL AI AGENT STORE INTEGRATION PLAYBOOK
Match documentation to the installed dependency before coding
Editorial recommendation · Reviewed 2026-10-31
When to choose this service
Choose Context7 when a coding agent repeatedly needs documentation across libraries. For one unsupported or unusual version, consult that release’s official documentation directly.
The maintained documentation corpus, indexing and retrieval service supplies external version context. A local MCP wrapper alone does not recreate that corpus or keep it current.
Suggested workflow
- Read the project’s dependency versions and identify the specific operation being implemented. Request a focused example with library and version hints.
- Inspect the returned library identity and source page. Resolve a missing or mismatched version before adopting the example.
- Adapt the example to the project and run a small behavior check. Keep the source link in the implementation notes when an API detail affects the design.
Acceptance checks before real work
- Confirm the example belongs to the intended library, not a similarly named package.
- Check whether version filtering actually selected the installed release.
- Exclude credentials and private application data from documentation queries.
If the workflow fails
If coverage is missing, use the exact release documentation or source repository. Narrow broad queries before consuming further calls, and stop or obtain a budget decision at the configured usage ceiling.