Context7

What is Context7?

Context7 is a documentation service for AI coding agents, built by Upstash. Models trained on older data often suggest outdated calls or invent APIs, so Context7 looks up the docs for the library version you are using and places the relevant snippets in the agent's context. Developers connect it as an MCP server, call it from the ctx7 CLI, or use the Search API and REST API directly. Teams can add a shared teamspace with access policies, per-member API keys and indexing of private GitHub repositories, so internal SDKs and component libraries become searchable too. Library owners can claim a library, configure it with a context7.json file and apply for verification. A companion product, Docs7, hosts documentation sites that are readable by agents. The Free plan includes 1,000 API calls per month, Pro is $10 per seat per month with 2,000 calls per seat, and Enterprise is custom priced.

Use cases

• Give coding agents current library docs
• Stop agents from inventing nonexistent APIs
• Index private repositories for a team
• Search library documentation through an API
• Publish agent-readable documentation sites
• Restrict which libraries agents may use

Who it's for

• Software developers
• Engineering teams
• Library maintainers
• Agent builders

Why use Context7

• Agents answer from the documentation for your library version instead of stale training data, which cuts down on broken code.
• A free account with an API key covers every public library, so individuals can start without paying.
• Pro and Enterprise accounts are never blocked at the included limit; extra calls are billed at $5 per 1,000.
• Teamspaces give each member a named key and the same access policies, so results stay consistent across a team.
• Only the lookup query and library name are sent to Context7, not your code or conversation history.

Context7 features

• Version-specific documentation and code examples for public libraries
• MCP server, CLI and REST API access
• Search API where each request counts as one API call
• Teamspaces with owner, admin and developer roles
• Named API key for each team member
• Private repository indexing on Pro and Enterprise
• Access policies that filter libraries, plus OAuth for MCP clients
• Library owner tools including context7.json and verification
• Context7 Ask endpoint for up to three websites on the Free plan
• SSO (SAML / OIDC) and self-hosted deployment on Enterprise

Context7 integrations

• GitHub
• Mintlify
• MCP clients

How to get started with Context7

1. Create a free account and an API key, then install Context7 as an MCP server or use the ctx7 CLI.
2. Ask your coding agent to use Context7 when a task involves a library.
3. The agent sends a short lookup with the library name and the question.
4. Context7 returns matching documentation and code examples for that version, and the agent writes code against them.
5. Teams can connect private GitHub repositories in the teamspace dashboard so internal libraries are indexed too.

Context7 limitations

• Free accounts are blocked after 1,000 API calls until the next month, apart from 20 bonus calls per day
• Private repositories are not available on the Free plan
• Private repository parsing is billed at $5 per million tokens
• Docs7 has its own plans, separate from Context7 plans

Context7 pricing

Yes, Context7 has a free plan. Paid plans start at $10 per month, according to the official pricing page on 24 Oct 2026.

Context7 plans

• Free

  $0

  • 1,000 API calls each month
  • Every public library
  • Library owner tools
  • Policies and OAuth
  • Up to 3 Context7 Ask sites

• Pro

  $10 per seat /month

  • 2,000 API calls per seat per month
  • $5 per 1,000 additional calls
  • Private repositories
  • Up to 10 teamspace members
  • Teamspace rules

• Enterprise

  Custom

  • SSO (SAML / OIDC)
  • SOC-2 compliance
  • Self-hosted deployment
  • Dedicated support
  • Unlimited teamspace members

Plans as listed on the official pricing page on 24 Oct 2026. Prices can change and may vary by country or exclude tax. Check the official pricing page.

Context7 alternatives

The main Context7 alternatives listed on Debuly are Mintlify, Exa, OpenWiki, Augment Code, Graft (now Trail-code) and OpenCode. The table compares pricing and platforms; order is A–Z.

1. 

   Augment Code

   Coding agent platform with a codebase Context Engine, Cosmos cloud agents and the Auggie CLI, priced as flat team plans plus usage.

   CodingAutomation and agentsDeveloper platforms
   PaidVisit Augment Code (official site)
2. 

   Exa

   Web search API for AI agents and apps that returns relevant passages with citations, billed pay-as-you-go from a free monthly credit.

   Developer platformsResearch and searchAutomation and agents
   FreemiumVisit Exa (official site)
3. 

   Graft (now Trail-code)

   Open source context layer that maps a codebase for coding agents and stores team learnings as Markdown files in git.

   CodingDeveloper platformsAutomation and agents
   Open sourceVisit Graft (now Trail-code) (official site)
4. 

   Mintlify

   Documentation platform for developer docs and API references, with an AI assistant, Git sync and a free Starter plan.

   Developer platformsWritingAutomation and agents
   FreemiumVisit Mintlify (official site)
5. 

   OpenCode

   Open-source AI coding agent for terminal, desktop and IDE, with optional low-cost Go model subscriptions and a team Console

   CodingDeveloper platformsAutomation and agents
   FreemiumVisit OpenCode (official site)
6. 

   OpenWiki

   LangChain command-line tool that generates and updates a linked Markdown wiki of your repository for coding agents and developers.

   CodingDeveloper platformsProductivity
   Open sourceVisit OpenWiki (official site)

Context7 FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from context7.com · fetched 23 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.