Qt4 Doc Mcp Server

io.github.jztanPublished by GitHub account jztanv0.7.0Updated Oct 24, 2026

Offline MCP server for local Qt 4.8, Qt 5, and Qt 6 documentation with full-text search.

VerifiedSTDIODesktop onlyKnowledge & MemoryDeveloper Tools

Overview

AI-generated overview

An offline MCP server that lets an assistant search and read a locally installed Qt 4.8, Qt 5, or Qt 6 HTML documentation set.

What you can do
Search the active Qt documentation set with full-text search, ranked results and highlighted snippets, with a configurable result limit (default 10, max 50). Read documentation pages converted to Markdown, optionally extracting a specific section or paginating long pages. Identify pages by their root-relative Markdown path rather than an online URL.
When to use it
Useful when you work on Qt 4.8, Qt 5, or Qt 6 code and want the assistant to look up classes, functions and guides from documentation already on your machine, without network access. It serves one documentation set at a time.
Requirements
A local process needing Python 3.11+ (or the published Docker image) and an existing Qt HTML documentation root set in QT_DOC_BASE. About 500 MB of disk for docs, cache and search index; SQLite with FTS5. QT_DOC_STATE_DIR is needed if the docs root is read-only. Defaults to HTTP on 127.0.0.1:8000, or stdio with a flag.
Before you install
The first start converts and indexes the documentation, which can take a minute or two and writes a Markdown cache and search index to disk. The server binds to 127.0.0.1 by default; changing SERVER_HOST to a wider address exposes it beyond the local machine. Qt documentation is under GFDL 1.3 and is served from your local copy.

Installation

In SourceWeft

  1. Open Qt4 Doc Mcp Server in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Qt Documentation MCP Server

mcp-name: io.github.jztan/qt4-doc-mcp-server

[PyPI Version] [License] [Python Version] [GitHub Issues] [CI] [Downloads]

Bring locally installed Qt 4.8, Qt 5, or Qt 6 documentation to your AI coding assistant. Works offline with one selected documentation set at a time.

Tool Reference | Changelog | Contributing | Troubleshooting

✨ Features

  • 🔌 Offline-First - Works entirely with local documentation
  • 🔍 Full-Text Search - Find what you need across all Qt docs
  • ⚡ Smart Caching - Fast responses for repeated queries
  • 🎯 Fragment Support - Extract specific sections when needed
  • 🛠️ MCP Standard - Compatible with Claude, VS Code, and other MCP clients

📦 Prerequisites

  • Python 3.11+ required (not needed when using the Docker image)
  • Qt HTML Documentation for one supported release (Qt 4.8, Qt 5, or Qt 6)
    • The included helper downloads Qt 4.8.4 only.
    • Point QT_DOC_BASE directly at an existing Qt 5/6 offline documentation root.
  • ~500MB disk space for docs + cache + search index
  • SQLite with FTS5 support (included in Python 3.11+ by default)

🚀 Installation

From PyPI (Recommended)

bash
pip install qt4-doc-mcp-server

From Source

bash
git clone https://github.com/jztan/qt4-doc-mcp-server.gitcd qt4-doc-mcp-serveruv sync --locked

Setup Qt Documentation

bash
# Automated setup (recommended, requires the cloned repo)python scripts/prepare_qt48_docs.py --segments 4
# This will:# - Download Qt 4.8.4 source archive# - Extract HTML documentation# - Create .env with sensible defaults# - Copy GFDL license file

The helper script ships with the repo, not the PyPI package. Installed from PyPI only? Download the docs manually instead:

bash
curl -LO https://download.qt.io/archive/qt/4.8/4.8.4/qt-everywhere-opensource-src-4.8.4.tar.gztar -xzf qt-everywhere-opensource-src-4.8.4.tar.gz qt-everywhere-opensource-src-4.8.4/doc/htmlmv qt-everywhere-opensource-src-4.8.4/doc/html ./qt4-docs-html

Then create a .env with QT_DOC_BASE pointing at that directory (see Configuration below).

Quick Start Commands

bash
# 1. Installpip install qt4-doc-mcp-server
# 2. Setup Qt docspython scripts/prepare_qt48_docs.py --segments 4
# 3. Build search index (optional; the server builds it on first start)qt-doc-build-index
# 4. Start serverqt-doc-mcp
# 5. Verify healthcurl -s http://127.0.0.1:8000/health

The legacy qt4-doc-mcp-server command remains available as an alias for existing client configurations.

Agent-friendly FTS CLI

After building the index, agents can search and receive materialized absolute Markdown paths:

bash
qt-doc-cli "accessible applications" --limit 5

The command reads the same .env settings as the server. Before searching, it automatically builds a missing/outdated FTS index and fully warms an incomplete Markdown cache. It then prints each result's title, absolute .md path, and FTS snippet to stdout; preparation messages, errors, and warnings go to stderr. Use qt-doc-warm-md --force after changing documentation in place.

🐳 Docker

Prebuilt multi-arch images (amd64/arm64) are published to GitHub Container Registry on every release. The container is offline-only: you mount your prepared Qt doc/html directory read-only at /docs, and all derived state (Markdown cache and search index) lives in a volume at /data.

bash
# Prepare Qt docs on the host first (one-time, writes to ./qt4-docs-html)python scripts/prepare_qt48_docs.py --segments 4
# Run from GHCRdocker run -d --name qt4-doc-mcp-server -p 8000:8000 \  -v "$PWD/qt4-docs-html:/docs:ro" \  -v qt4-doc-data:/data \  ghcr.io/jztan/qt4-doc-mcp-server:latest
# Verifycurl -s http://127.0.0.1:8000/health

Using only the published image, without cloning this repo? Download the Qt 4.8.4 docs directly instead of running the prepare script:

bash
curl -LO https://download.qt.io/archive/qt/4.8/4.8.4/qt-everywhere-opensource-src-4.8.4.tar.gztar -xzf qt-everywhere-opensource-src-4.8.4.tar.gz qt-everywhere-opensource-src-4.8.4/doc/htmlmv qt-everywhere-opensource-src-4.8.4/doc/html ./qt4-docs-html

Any existing Qt 5 or Qt 6 doc/html tree (from the Qt installer or distro documentation packages) works as the mount source too.

First start converts and indexes the documentation into the /data volume; subsequent starts reuse it. Depending on the docset size, the first start can take a minute or two before the health endpoint responds.

Docker Compose

With docs in the default ./qt4-docs-html location, no configuration is needed:

bash
docker compose up -d

For docs elsewhere (including Qt 5 or Qt 6 doc/html trees, which the server also supports), point QT_DOC_HTML_PATH at them. If port 8000 is taken on the host, set HOST_PORT in the same file (with plain docker run, change the left side of -p instead):

bash
cp .env.docker.example .env.docker   # set QT_DOC_HTML_PATHdocker compose --env-file .env.docker up -d

Or use the convenience script, which checks Docker is running, creates .env.docker on first run, builds and starts the service, and waits for the health endpoint:

bash
./deploy.sh

Point your MCP client at http://127.0.0.1:8000/mcp (streamable HTTP). Qt documentation is licensed under GFDL 1.3; the container serves your local copy and never redistributes it.

⚙️ Configuration

These settings apply to native (pip or source) runs. In Docker they are already set inside the container (QT_DOC_BASE=/docs, QT_DOC_STATE_DIR=/data, SERVER_HOST=0.0.0.0); the QT_DOC_HTML_PATH variable in .env.docker is not a server setting, just the host path mounted at /docs.

Create a .env file in the repo root. The helper script writes sensible defaults; adjust as needed:

VariableDefaultPurpose
QT_DOC_BASErequiredAbsolute path to one Qt 4.8, Qt 5, or Qt 6 HTML documentation root. The server detects the active docset.
QT_DOC_STATE_DIR$QT_DOC_BASE/.indexOptional writable directory for the FTS index and Markdown cache. Use this when the documentation root is read-only.
PREINDEX_DOCStrueBuild search index automatically at startup if not present.
PRECONVERT_MDfalseWarm the Markdown cache automatically at MCP startup.
SERVER_HOST127.0.0.1Bind address for the FastMCP server (0.0.0.0 for containers).
SERVER_PORT8000TCP port for streamable HTTP transport.
MCP_LOG_LEVELWARNINGLogging verbosity (DEBUG/INFO/WARNING/ERROR).
MD_CACHE_SIZE512In-memory CachedDoc LRU capacity (counts pages).
DEFAULT_MAX_MARKDOWN_LENGTH20000Default maximum characters returned per request (prevents token limit issues).

The tools identify documents by their exact root-relative Markdown path, not an online URL. For example, use qcompleter.md for a Qt 4 page, qtdoc/accessible.md for a Qt 5/6 global page, or qtcore/qobject.md for a Qt 5/6 Core page. By default, each docset stores its own index and Markdown cache under $QT_DOC_BASE/.index/, so switching QT_DOC_BASE reuses its existing derived state. Set QT_DOC_STATE_DIR to relocate both to a writable directory. The Markdown cache mirrors the documentation tree: for example, qtcore/qobject.md is cached as .index/md/qtcore/qobject.md plus qobject.meta.json with the default state directory.

🔌 MCP Client Setup

By default, the server exposes an HTTP endpoint at http://127.0.0.1:8000/mcp. Register it with your preferred MCP-compatible agent using the instructions below.

Stdio transport

Run the server over stdio instead of HTTP with:

bash
qt-doc-mcp --transport stdio

For stdio-only MCP clients, configure that command with args: ["--transport", "stdio"]. Startup indexing and Markdown-cache progress are written to stderr, leaving stdout exclusively for MCP protocol messages.

Visual Studio Code (Native MCP Support)

VS Code has built-in MCP support via GitHub Copilot (requires VS Code 1.102+).

Using CLI (Quickest):

bash
code --add-mcp '{"name":"qt-docs","type":"http","url":"http://127.0.0.1:8000/mcp"}'

Using Command Palette:

  1. Open Command Palette (Cmd/Ctrl+Shift+P)
  2. Run MCP: Open User Configuration (for global) or MCP: Open Workspace Folder Configuration (for project-specific)
  3. Add the configuration:
    json
    {  "servers": {    "qt-docs": {      "type": "http",      "url": "http://127.0.0.1:8000/mcp"    }  }}
  4. Save the file. VS Code will automatically load the MCP server.

Manual Configuration: Create .vscode/mcp.json in your workspace (or mcp.json in your user profile directory):

json
{  "servers": {    "qt-docs": {      "type": "http",      "url": "http://127.0.0.1:8000/mcp"    }  }}
Claude Code

Add to Claude Code using the CLI command:

bash
claude mcp add --transport http qt-docs http://127.0.0.1:8000/mcp

Or configure manually in your Claude Code settings file (~/.claude.json):

json
{  "mcpServers": {    "qt-docs": {      "type": "http",      "url": "http://127.0.0.1:8000/mcp"    }  }}
Codex CLI

Add to Codex CLI using the command:

bash
codex mcp add qt-docs -- npx -y mcp-client-http http://127.0.0.1:8000/mcp

Or configure manually in ~/.codex/config.toml:

toml
[mcp_servers.qt-docs]command = "npx"args = ["-y", "mcp-client-http", "http://127.0.0.1:8000/mcp"]

Note: Codex CLI primarily supports stdio-based MCP servers. The above uses mcp-client-http as a bridge for HTTP transport.

Kiro

Kiro primarily supports stdio-based MCP servers. For HTTP servers, use an HTTP-to-stdio bridge:

  1. Create or edit .kiro/settings/mcp.json in your workspace:
    json
    {  "mcpServers": {    "qt-docs": {      "command": "npx",      "args": [        "-y",        "mcp-client-http",        "http://127.0.0.1:8000/mcp"      ],      "disabled": false    }  }}
  2. Save the file and restart Kiro. The active Qt documentation tools will appear in the MCP panel.

Note: Direct HTTP transport support in Kiro is limited. The above configuration uses mcp-client-http as a bridge to connect to HTTP MCP servers.

Generic MCP Clients

Most MCP clients use a standard configuration format. For HTTP servers:

json
{  "mcpServers": {    "qt-docs": {      "type": "http",      "url": "http://127.0.0.1:8000/mcp"    }  }}

For clients that require a command-based approach with HTTP bridge:

json
{  "mcpServers": {    "qt-docs": {      "command": "npx",      "args": ["-y", "mcp-client-http", "http://127.0.0.1:8000/mcp"]    }  }}

🛠️ Available Tools

The server provides 2 MCP tools for working with the active local Qt documentation set:

  1. read_documentation - Read and convert pages from the active Qt documentation set to Markdown

    • Fragment extraction (#details, #public-functions)
    • Pagination with start_index and max_length
    • Section-only mode for targeted content
    • Returns Markdown with normalized links and GFDL attribution
  2. search_documentation - Full-text search across the active Qt documentation set

    • SQLite FTS5 with BM25 relevance ranking
    • Context snippets with highlighted matches
    • Configurable result limits (default: 10, max: 50)

For detailed API documentation including parameters, return values, examples, and error handling, see the Tool Reference.

📚 Related Resources

📄 License

  • Code: MIT License (see LICENSE).
  • Qt Documentation: © The Qt Company Ltd. and contributors, licensed under GFDL 1.3. This server converts locally obtained docs and includes attribution in outputs. If you redistribute a local mirror, include LICENSE.FDL and preserve notices.
  • See THIRD_PARTY_NOTICES.md for more details.

Source: README.md at commit 9998403

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.7.0LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts