M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
C

Context

Community
Local-first documentation for AI agents — offline, instant, private doc retrieval.
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #176 of 438
Stars
★ 422 Very popular
Transport
stdio (local process) · Streamable HTTP
Runtime
Node.js · Docker
Credentials
No credential needed
License
Apache-2.0
Last commit
Tools
4
56FMRS · C

Context is a practical MCP server that provides local, reliable documentation retrieval for AI agents. It solves the common problem of AI using outdated docs and simplifies setup via a community registry. For developers, it's a time-saver and accuracy booster.

Strongest · Documentation 15/20 Weakest · Reliability 8/20

Reliability
8/20
Security and permissions
10/20
Maintenance
9/20
Documentation
15/20
Setup experience
14/20
Why each score
Reliability 8/20
The manifest and README present a coherent design (stdio/HTTP MCP server, registry-backed package download, SQLite/FTS5 local query), but no source code, CI workflows, or committed tests were supplied to verify startup, handshake, or tool behavior. README claims were treated as untrusted evidence, and the full tool list is not specified. Deductions for lack of execution evidence and thin error-path documentation yield 8.
Security and permissions 10/20
No red-line issue is evident from the supplied material: credentials are stored with 0600 permissions and domain-scoped matching, and doc queries are local. However, first-use auto-downloads registry packages without documented checksum/signature verification or explicit confirmation, least-privilege scoping of the MCP session is not demonstrated, and no encryption-at-rest is mentioned. Main risks are visible but confirmation and boundary controls are incomplete, so 10.
Maintenance 9/20
The repository is not archived, has an Apache-2.0 license, a released npm package (0.2.3), and 0 open issues. There is no commit history, release cadence, dependency-update evidence, or security-response channel, so maintenance governance is under-evidenced. Stars are discovery signals only and were not scored. Score 9.
Documentation 15/20
The README is layered and strong: quick-start configs for many clients, full CLI reference, auth storage details, Docker, architecture diagram, FAQ, and registry contribution guide. It lacks a dedicated MCP tool parameter/reference section and a troubleshooting/limitations section, so it is not a perfect 20. Score 15.
Setup experience 14/20
Setup is straightforward: a single global npm install plus an MCP client config entry, with copy-paste examples for major clients and a Docker/HTTP alternative. The README does not state Node/npm version prerequisites, and there is no verifiable CI or committed-test evidence, so per static calibration it is capped at 15; I score 14.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessReads local filesUses the networkConnects to a database

Best for

  • Developers who need fast, reliable, and private documentation retrieval
  • Developers using popular libraries like Next.js, React, Prisma
  • AI agent users in offline or low-latency environments
  • Teams that need to share custom documentation packages

Not for

  • Users who need automatic updates for all packages (versions are fixed and need manual updates)
  • Teams that prefer not to install through a global package manager
  • Users expecting a cloud-hosted service

Required permissions

  • Read/write to ~/.context directory (stores docs and auth data)
  • Network access (for first-time package downloads and registry queries)
  • Communication with clients via stdio or HTTP

Risks and side effects

  • Registry packages may contain outdated or inaccurate content (community-maintained)
  • Installed packages consume disk space (a few MB each)
  • In HTTP mode, if not restricted to localhost, could be exposed to other clients on the network

Setup

Before you start

Runtime：Node.js · Docker

  1. Install globally: npm install -g @neuledge/context
  2. Configure the MCP server for your AI client (e.g., Claude Desktop config file)
  3. Restart the client and start asking questions.
.mcp.json 
{"mcpServers":{"context":{"command":"context","args":["serve"]}}}

Shown for Claude Code. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "context": {
      "command": "context",
      "args": [
        "serve"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add context -- context serve

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After installing, the client's tool list should show search_packages, download_package, get_docs, and list_packages; alternatively ask "How do I create middleware in Next.js?" — an answer citing locally downloaded docs confirms the connection.

Troubleshooting

  1. Ensure it's installed globally: npm install -g @neuledge/context
  2. Check the command and arguments in your client's MCP config
  3. Verify network connectivity for first-time downloads
  4. Check that ~/.context directory exists and is writable
  5. Run context list to verify packages are installed

Things to try

Once connected, you can ask your AI assistant things like:

  • How do I create middleware in Next.js?
  • Look up the ToolLoopAgent API in AI SDK v6
  • List my locally installed documentation packages
  • How do I define a one-to-many relation in Drizzle ORM?

Tools 4

search_packages read-only
Search the community registry for available documentation packages.
download_package writes
Download and install a pre-built documentation package from the registry.
get_docs read-only
Retrieve documentation sections relevant to a topic from installed packages.
list_packages read-only
List locally installed documentation packages.

Use cases

Get up-to-date, version-specific library docs while developing
Access documentation offline or in restricted networks
Avoid AI providing wrong answers due to outdated docs
Build docs packages for private or internal libraries
Build packages from websites using llms.txt

Supported clients

Claude Code
Claude Desktop
Cursor
Codex
VS Code
Windsurf
Zed
Goose
OpenCode

Listed from the project's documentation, not tested by this site.

Overview

Context is an MCP server backed by a community-driven package registry with 100+ pre-built documentation packages for popular libraries. When your AI agent needs documentation, it searches the registry, downloads the right package, and queries it locally — all automatically. Install once, configure once, then just ask your AI. Docs are stored as compact SQLite databases in ~/.context/packages/, making subsequent queries entirely local.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision 4a20dca68823 Data synced 2026-10-11 Read the FMRS scoring method