← Back to directory
C

Context7

Official
Upstash's official server providing up-to-date third-party library docs for AI coding assistants
GitHub source repository ↗
★ 59.7k Stars Category · Dev Tools Very popular
80FMRS · B
Source and config reviewed
Reliability
14/20
Security and permissions
16/20
Maintenance
17/20
Documentation
15/20
Setup experience
18/20

Officially maintained by Upstash; the tool design is minimal (just two read-only tools) and addresses a specific, common pain point — an AI assistant coding against stale API knowledge. Low install barrier, usable without an API key. This review is a static check of source and docs, not an actual call test.

Read the scoring and verification method →
Officially maintained by Upstash, this MCP server addresses a common problem: an AI coding assistant's training data going stale, leading it to suggest outdated API usage. Two tools first resolve a library name into a Context7 library ID, then retrieve documentation for that library's current version — so the assistant references today's real API instead of what it memorized during training.

Tools

resolve-library-id
Resolve a general library name into a Context7-compatible library ID
query-docs
Retrieve up-to-date documentation for a library using its Context7 library ID

Setup

Start via `npx -y @upstash/context7-mcp` and use it immediately without an API key (subject to rate limits); apply for a CONTEXT7_API_KEY at context7.com for higher quota and set it as an env var. An MCPB one-click install bundle and a remote Streamable HTTP endpoint (`https://mcp.context7.com/mcp`) are also available as alternatives.
claude_desktop_config.json
{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp"],
      "env": {
        "CONTEXT7_API_KEY": "<OPTIONAL_API_KEY>"
      }
    }
  }
}

Fit and risk

Best for

  • Developers using fast-moving frameworks/libraries worried about the AI suggesting stale code
  • Scenarios wanting zero-config documentation lookup

Not for

  • Looking up internal/private codebase documentation (Context7 targets publicly published open-source libraries)
  • Cases needing very high coverage of obscure, niche libraries (coverage depends on what Context7's platform has indexed)

Required permissions

  • Usable without an API key (subject to a free-tier rate limit); CONTEXT7_API_KEY is an optional credential for higher quota
  • Read-only documentation lookup — no code execution or local filesystem access involved

Risks and side effects

  • The free tier has limited quota — high-frequency use may hit rate limits
  • Documentation content comes from Context7's platform index, so its accuracy and freshness depend on that platform's crawl cadence

Troubleshooting

  1. Empty search results: verify the library name resolves correctly with resolve-library-id first
  2. Hitting rate limits: apply for a CONTEXT7_API_KEY for higher quota
  3. Remote endpoint connection fails: check network access to mcp.context7.com

Use cases

Have the AI assistant look up a library's current, correct usage while coding, avoiding deprecated APIs
Quickly get official documentation highlights for an unfamiliar third-party library
Verify during code review that an API call matches the library's latest recommended usage

Supported clients

Claude CodeFull support
VS CodeFull support
CursorFull support
ClineFull support
AmpFull support