Installation

Requirements

• Python 3.11, 3.12, or 3.13
• pipx (recommended) or pip

Install with pipx (recommended)

pipx installs Docmancer in an isolated environment so it does not interfere with other Python packages:

pipx install docmancer --python python3.13

If pipx picks an unsupported interpreter, pin one explicitly with --python python3.13.

Install with pip

pip install docmancer

Optional extras

The base install gives you SQLite FTS5 lexical retrieval and the MCP runtime. The extras below pull in heavier dependencies only when you need them:

You can combine extras:

pip install 'docmancer[local,vector]'

Verify the installation

docmancer doctor

doctor reports config status, SQLite index health, loader availability, Qdrant status, embeddings provider, vector / lexical drift, and installed agent skills.

First-run setup

Run the setup wizard to create your config, initialize local storage, and connect coding agents:

docmancer setup

Setup will:

1. Create ~/.docmancer/docmancer.yaml with default settings.
2. Initialize the SQLite database at ~/.docmancer/docmancer.db.
3. Auto-detect installed coding agents (Claude Code, Cursor, Codex, Cline, Gemini, OpenCode, Claude Desktop, GitHub Copilot).
4. Install skill or instruction files for docs retrieval.
5. Register the local docmancer mcp serve runtime for agents that support MCP configuration.

You can also install all agent integrations non-interactively, or only specific agents:

docmancer setup --all
docmancer setup --agent claude-code --agent cursor

First ingest and Qdrant binary download

On the first run of docmancer ingest or docmancer add, Docmancer downloads:

• The pinned Qdrant binary (~60 MB) from the v1.14.1 GitHub release into ~/.docmancer/qdrant/.
• The FastEmbed models (~500 MB) for dense + sparse vectors into ~/.docmancer/embeddings-cache/.

After that, ingest and query are offline.

If your platform has no matching Qdrant binary, Docmancer transparently falls back to sqlite-vec for dense vectors. You can also skip vectors entirely:

docmancer ingest ./docs --no-vectors

Or set DOCMANCER_AUTO_VECTORS=0 to keep vectors off globally.

Install an API MCP pack

After setup, install a version-pinned pack:

docmancer install-pack open-meteo@v1
docmancer mcp doctor

Open-Meteo is keyless, so mcp doctor reports no required credentials. After that, ask your agent to get the current weather in New York Central Park. The agent calls open_meteo__v1__forecast with GET /v1/forecast?latitude=40.7812&longitude=-73.9665&current_weather=true.

Where things live