Distillery¶
Team Knowledge, Distilled
Distillery is a team knowledge base accessed through Claude Code skills. It refines raw information from working sessions, meetings, bookmarks, and conversations into concentrated, searchable knowledge — stored as vector embeddings in DuckDB and retrieved through natural language.
Runs locally over stdio or as a hosted HTTP service with GitHub OAuth for team access.
Who is Distillery for?¶
- Developers and teams who use Claude Code and want to capture, organize, and retrieve knowledge without leaving their workflow.
- Team leads and operators who want a shared knowledge base their team can access through a hosted MCP server with GitHub authentication.
- Anyone who wants to stop losing context — capture decisions, insights, and references as they happen, and retrieve them through natural language when you need them.
Skills¶
Distillery provides 15 Claude Code slash commands:
| Skill | Purpose | Example |
|---|---|---|
/distill |
Capture session knowledge with dedup detection | /distill "We decided to use DuckDB for local storage" |
/recall |
Semantic search with provenance | /recall distributed caching strategies |
/pour |
Multi-entry synthesis with citations | /pour how does our auth system work? |
/compass |
Internal vs ambient directional assessment | /compass agent eval harnesses |
/bookmark |
Store URLs with auto-generated summaries | /bookmark https://example.com/article #caching |
/minutes |
Meeting notes with append updates | /minutes --update standup-2026-03-22 |
/classify |
Classify entries and triage review queue | /classify --inbox |
/watch |
Manage monitored feed sources | /watch add github:duckdb/duckdb |
/radar |
Ambient feed digest with source suggestions | /radar --days 7 |
/tune |
Adjust feed relevance thresholds | /tune --digest 0.40 |
/setup |
Onboarding wizard for MCP connectivity and config | /setup |
/digest |
Team activity summary from internal entries | /digest --days 7 |
/gh-sync |
Sync GitHub issues/PRs into the knowledge base | /gh-sync owner/repo |
/investigate |
Deep context builder with relationship traversal | /investigate authentication flow |
/briefing |
Knowledge dashboard (solo and team mode) | /briefing --project distillery |
Quick Start¶
Step 1: Install the Plugin¶
This installs all 15 skills. The plugin does not configure an MCP server automatically — run /setup to add one. The recommended setup runs locally via uvx --from 'distillery-mcp[fastembed]>=0.6.0' distillery-mcp — a private, self-contained knowledge base on your machine, with on-device fastembed embeddings (no API key required). Requires Python 3.11+ and uv.
Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh
Step 2 (Optional): Use Jina or OpenAI Instead¶
The default fastembed provider runs offline with no API key. To use a hosted embedding service instead, set DISTILLERY_EMBEDDING_PROVIDER and the matching API key in your shell before launching Claude Code:
# Jina (free tier at jina.ai)
export DISTILLERY_EMBEDDING_PROVIDER=jina
export JINA_API_KEY=jina_...
# Or OpenAI
export DISTILLERY_EMBEDDING_PROVIDER=openai
export OPENAI_API_KEY=sk-...
Restart Claude Code and run /setup to complete onboarding.
Try the Hosted Demo (Opt-In)¶
Want to evaluate without installing locally? Configure the hosted demo at distillery-mcp.fly.dev instead of a local server:
Demo Server
distillery-mcp.fly.dev is a demo server for evaluation only. Do not store sensitive or confidential data.
See Local Setup for full configuration (embedding providers, cloud storage, etc.) or deploy your own instance for team use.
License¶
Apache 2.0 — see LICENSE for details.