Skip to main content
Glama
rkratky

DocsHub MCP Server

by rkratky
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
DOCSHUB_CONFIGNoLocal file path or https:// URL to the config file. Overrides other config sources.
DOCSHUB_CACHE_TTLNoSeconds to cache fetched documentation in memory. Set to 0 to disable caching.1800
DOCSHUB_CONFIG_RELOAD_INTERVALNoSeconds between config reload checks. Set to 0 to disable polling.900

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}
logging
{}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
completions
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
list_available_docsA

List all configured documentation projects.

Always call this first to discover available project names before using get_project_docs. Pass the exact value of the name field to get_project_docs.

Returns: List of dicts with name and description keys. Returns an empty list if no projects are configured.

get_project_docsA

Fetch documentation for a project.

Automatically tries llms-full.txt first, which returns the complete documentation in a single response — no further calls needed. If llms-full.txt is unavailable, falls back to llms.txt and returns an index of URLs; use read_doc_page with those URLs to retrieve individual pages.

Call list_available_docs first to get valid project names.

read_doc_pageA

Fetch the Markdown content of a specific documentation page.

Use URLs obtained from get_project_docs when it returns an llms.txt index. Do not call this if get_project_docs already returned complete documentation. Only URLs from configured project hosts are permitted.

Prompts

Interactive templates invoked by user choice

NameDescription
query_docsAnswer a question using a project's documentation. Guides the model through the full lookup workflow: fetch docs, then answer.
summarize_projectSummarize the documentation for a project. Guides the model to fetch and distil the key topics and structure.

Resources

Contextual data attached and managed by the client

NameDescription
projects_resourceRegistry of all configured documentation projects. Returns a JSON array with ``name`` and ``description`` for each project. Use the exact ``name`` value when calling ``get_project_docs`` or reading the ``docshub://project/{name}/docs`` resource. Returns: JSON-encoded list of project dicts with ``name`` and ``description``.
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.9/5.0

Scored across 3 tools

Disambiguation5/5

Each tool has a distinct, non-overlapping role in the documentation workflow: listing projects, fetching project docs (full or index), and reading individual pages. The descriptions clearly define when to use each tool.

Naming Consistency5/5

All tool names follow a consistent snake_case verb_noun pattern: list_available_docs, get_project_docs, read_doc_page. This is predictable and easy for an agent to interpret.

Tool Count5/5

With 3 tools, the server is tightly scoped to its purpose of retrieving documentation. Each tool is essential to the workflow, and there is no redundancy or excess.

Completeness5/5

The tool set covers the full lifecycle of fetching documentation: discover projects, retrieve either full content or an index, and then fetch specific pages. The fallback mechanism ensures no dead ends for users.

Maintenance

ActivitySlowing
ResponsivenessNo issues

Resources