Atlas / MCP servers / andrewyng / Context Hub

Context HubCAUTION

Name: Context Hub
Author: andrewyng

mcp/andrewyng/context-hub

MCP server by andrewyng · Audited 2026-10-12

⚑ Report / re-scan Source ↗

Verdict

CAUTION

Grade

Trust score

47 /100

Exposed tools

23 21r · 1w · 1d

Transport

stdio

License

MIT

Stars

13,975

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

Coding agents hallucinate APIs and forget what they learn in a session. Context Hub gives them curated, versioned docs, plus the ability to get smarter with every task. All content is open and maintained as markdown in this repo — you can inspect exactly what your agent reads, and contribute back.

[](LICENSE) [](https://www.npmjs.com/package/@aisuite/chub) [](https://nodejs.org)

Quick Start

npm install -g @aisuite/chub
chub search openai                 # find what's available
chub get openai/chat --lang py     # fetch current docs (Python version)

How It Works

Chub is designed for your coding agent to use (not for you to use!). You can prompt your agent to use it (e.g., "Use the CLI command chub to get the latest API documentation for calling OpenAI. Run 'chub help' to understand how it works.") Or by creating an agent skill to use Chub using SKILL.md, and ideally prompting your agent to remember to use this skill. (If you are using Claude Code, create the directory ~/.claude/skills/get-api-docs and put SKILL.md there.)

Most of the time, it's simple — search, fetch, use:

chub search "stripe payments"        # find relevant docs
chub get stripe/api --lang js        # fetch the doc
# Agent reads the doc, writes correct code. Done.

When the agent discovers a gap, it can annotate locally for next time:

chub annotate stripe/api "Needs raw body for webhook verification"

# Next session, chub get notes that an annotation exists. Pass --with-annotations
# to include it — annotations are treated as untrusted input by default.

Feedback flows back to authors — chub feedback stripe/api up or down — vote the docs up or down so they can get better for everyone over time.

Conten

Read from source at commit 9628ef7add72OBSERVED · 2026-10-12

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.

claude-code

claude mcp add context-hub -- npx -y context-hub

claude-desktop

{
  "mcpServers": {
    "context-hub": {
      "command": "npx",
      "args": [
        "-y",
        "context-hub"
      ]
    }
  }
}

Exposed tools (23)

21 read · 1 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

Tool	Risk	Description
`Alpha`	read	Alpha topic
`Beta`	read	Beta topic
`Next.js`	read	React framework for production
`OpenAPI`	read	OpenAPI tooling docs
`Playwright`	read	Browser automation library
`React`	read	React UI library
`api`	read	Payment processing platform
`auth`	read	User authentication with social login and session management
`cache`	read	In-memory data store
`chat`	read	Chat completions API
`chub_annotate`	destructive	Read, write, clear, or list agent annotations. Modes: (1) list=true to list all, (2) id+note to write, (3) id+clear=true to delete, (4) id alone to read. Annotations persist locally across sessions.
`chub_feedback`	write	Send quality feedback (thumbs up/down) for a doc or skill to help authors improve content
`chub_get`	read	Fetch the content of a doc or skill by ID from Context Hub
`chub_list`	read	List all available docs and skills in Context Hub. Alias of
`chub_search`	read	Search Context Hub for docs and skills by query, tags, or language
`driver`	read	Document database driver for storing and querying JSON data
`errors`	read	Error monitoring and performance tracking for production apps
`eslint-config-next`	read	ESLint preset for Next.js apps
`identity`	read	Authentication and authorization platform with SSO and MFA
`messaging`	read	SMS and messaging API for sending notifications
`node-fetch`	read	Fetch API for Node.js
`payments`	read	Payment processing for commerce
`testing`	read	Browser automation for end-to-end testing of web applications

Trust audit

CAUTIONgrade F · trust 47/100 Install with care. The audit found things worth knowing before you trust its output.

Layer	What it checks	Result
L0	Provenance & inventory	PASS
L1	Static analysis of the code	WARN
L2	Instruction surface (what it tells the agent)	PASS
L3	Class-specific surface	WARN
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (1 observation(s))
Network: declared (4 observation(s))
Shell: declared (2 observation(s))
Dependencies: not all pinned
Secrets in source: found

Findings (25)

MEDIUMHard-coded secrets · secret.db_uri · CWE-798, CWE-321

content/apache-airflow/docs/providers-celery/python/DOC.md:83

export AIRFLOW__CELERY__RESULT_BACKEND="db+postgresql://airflow:secret@localhost:5432/airflow"

MEDIUMHard-coded secrets · secret.db_uri · CWE-798, CWE-321

content/apache-airflow/docs/providers-mysql/python/DOC.md:47

export AIRFLOW_CONN_MYSQL_DEFAULT='mysql://airflow:airflow@mysql:3306/analytics?charset=utf8mb4'

MEDIUMHard-coded secrets · secret.db_uri · CWE-798, CWE-321

content/asyncpg/docs/package/python/DOC.md:80

"postgresql://app:[email protected]:5432/appdb"

MEDIUMHard-coded secrets · secret.db_uri · CWE-798, CWE-321

content/dagster/docs/celery/python/DOC.md:57

export DAGSTER_CELERY_BROKER_URL="amqp://guest:guest@localhost:5672//"

MEDIUMHard-coded secrets · secret.db_uri · CWE-798, CWE-321

content/django/docs/environ/python/DOC.md:74

DATABASE_URL=postgres://app:[email protected]:5432/app