Context7
Current Library Docs for AI Coding — What Context7 Is, and Where It Fits.
AI assistants answer from a training snapshot; your dependencies move on. Context7 retrieves current, version-exact library documentation at answer time — through an MCP server in your editor, a REST API or SDK on your server, or a browsable catalog. This page lays out the surfaces, the real scale of the index and the limits, with the figures dated.
Comparison basis
Where these figures come from
Every number on this page comes from VUST's own crawl of Context7's public sitemap, catalog and API on 2026-08-02: 106,573 documentation sources across 23 sitemap files, and a live top-50 rankings page that reduces to 42 product families once same-product entries are merged. The crawl's caveat is kept with the numbers: these are sources, not distinct products. Indexes move — verify on Context7's site before relying on coverage.
Unless a separate method is stated above, this page does not claim a head-to-head product-quality benchmark.
Checked: 2026-08-02
See the difference
Three surfaces, one index — the line between them is who is in the loop.
02·Practical use cases
Who reaches for Context7
Developers in MCP-capable editors
Stop the assistant from inventing APIs from stale training data
The MCP server feeds current, version-exact docs into the session as you work.
Teams pinning dependency versions
Answer against the version you actually run, not the latest
Library plus version is the query unit — the index resolves that exact slice of documentation.
Products needing docs server-side
Build documentation-aware features without scraping
The REST API and TypeScript SDK expose the same index programmatically.
03·How it works
How a Context7 lookup works
MCP server when a person works interactively in an editor or agent; REST API or SDK when your own software needs the lookup; the web catalog to check coverage first.
Name collisions are real — Bolt the ride-hailing app and Bolt.new the app builder are different entries. Resolve the library and pin the version before trusting an answer.
The retrieved documentation slice rides along with your question, so the answer reflects the current API surface rather than the model's training snapshot.
04·Same tool · in Telegram
Sharpen the question first
@vustbot · Draft the spec, the failing case and the exact question with @vustbot, then take it to your docs-aware tooling. Free tier, no card.
05·Quality & trust
Context7, honestly
An index of sources, not an oracle
The 106,573 entries are documentation sources — repositories, websites, llms.txt files, OpenAPI documents — and several can describe one product. Source trust and entity resolution stay your job.
Privacy is a working rule, not a toggle
Context7's own data-privacy documentation describes query distillation and sanitization. Send the question, never secrets or code you cannot share — that rule holds for any third-party retrieval service.
Figures move
Scale, plans and rankings here were read from Context7's public surfaces on 2026-08-02. Verify on its site before you build against an assumed quota.
Frequently asked questions
What is Context7?
An open-source documentation service that feeds current, version-exact library documentation to AI tools. Models are trained on a snapshot and confidently invent APIs that have since changed; Context7 narrows that gap by retrieving the documentation for the exact library and version you name, at answer time. It ships as an MCP server for editors and agents, a REST API and a TypeScript SDK for server-side use, and a browsable web catalog.
What do the 106,573 sources actually count?
Documentation sources, not products: on 2026-08-02 the public sitemap resolved to 106,573 entries across 23 sitemap files — 75,050 repository or source namespaces, 28,625 websites, 2,312 llms.txt files and 586 OpenAPI documents. Several sources can describe one product, and mirrors exist, so treat the number as index scale, not as a count of distinct libraries.
MCP server, CLI, SDK or API — which surface fits?
MCP when a person is in the loop: an MCP-capable editor or agent resolves docs interactively as you work. REST API or the TypeScript SDK when software is in the loop: your own service needs documentation retrieval as a feature. The web catalog is for checking coverage and disambiguating same-named entries before you rely on either.
Is it free?
A free plan exists, and paid plans are listed on Context7's own pricing page. Plan compositions and limits change, so verify there before building anything on an assumed quota — this page deliberately quotes no prices.
What about privacy and secrets?
Context7's data-privacy documentation (checked 2026-08-02) describes query distillation and sanitization on their side. The working rule stands regardless of any vendor's policy: never put secrets, credentials or code you cannot share into a query to a third-party retrieval service. Send the question, not the codebase.
Does VUST use Context7 — can @vustbot answer from current docs?
Not today, and this page says so plainly. @vustbot is a multi-model coding chat: it answers from model knowledge plus whatever you paste, it does not fetch version-pinned documentation, and it holds no MCP authority. A current-docs mode — library@version plus your question, answered with citations — is PLANNED, and this paragraph changes only after that mode ships and passes acceptance.
Ready when you are
Context7 keeps the docs current; the surface depends on who is in the loop.
MCP for interactive editor work, REST or SDK for server-side features, the catalog for coverage checks. Checked 2026-08-02 — verify plans and coverage on Context7's site.