Ullage
БесплатноНе проверенManage your wine cellar through natural language with your AI agent, including adding wines from label or receipt photos, updating inventory, logging consumptio
Описание
Manage your wine cellar through natural language with your AI agent, including adding wines from label or receipt photos, updating inventory, logging consumption with ratings, and exporting data.
README
Your wine cellar, run by your AI agent — OpenClaw, Hermes, or any MCP client.
What it does
- Chat with your agent to manage your cellar: add, list, update, consume (rating + notes), hold, release, recommend, export.
- Share a wine label or receipt photo — your agent reads it and saves confirmed wines in one batch. ullage runs no OCR; your agent's own vision does it.
- Idempotent confirmed writes — repeated chat/button actions can use the same idempotency key without creating duplicates.
- Hold protection — mark bottles that should not be opened yet; ordinary consumption and recommendations respect the hold.
- Consumption history — every bottle opened is logged with date, rating, and notes.
- You own the data — SQLite is the default adapter (
~/.ullage/cellar.db), export anytime. OSS, MIT.
Install — tell your agent
Give OpenClaw / Hermes / any MCP client this one line:
Check https://github.com/nukk-pain/ullage and install it.
The agent reads docs/agent-install.md, installs the MCP server for your runner (SQLite default adapter — no token, no Docker), starts a fresh session, and then manages your cellar per that runbook.
Agent reading this repo: follow docs/agent-install.md to install and then operate the cellar. You need a shell with
gitand Node.js 20+.
Then just talk:
- "Add a Barolo 2016." — or share a wine label photo; it reads the label and adds the bottle.
- Share a receipt photo — it confirms the wine rows, saves all of them atomically, captures price / purchase date / store, and skips non-wine items.
- "I drank the Chablis tonight — 2/5, too light." — logged with your rating and note.
- "Hold the Barolo for our anniversary." — marked so it will not be opened accidentally.
- "What should I drink with fried chicken?" — it recommends from your cellar.
How it works
ullage is the system of record — a wine cellar your agent reads and writes through MCP tools. The intelligence (reading labels, pairing, "what should I drink") is your agent's own LLM working over that data. By default there is no server, no token, and no Docker — just a local SQLite adapter file you own.
The public install path is intentionally independent of any external wine API. Experimental alternate storage adapters are selected only by environment variable and are not needed for normal agent setup.
Optional safety mode: set ULLAGE_APPEND_ONLY=true with SQLite to prevent direct edits to saved
wine identity/purchase fields. Later changes should be recorded as consume, hold, release, or note
events.
License
MIT — see LICENSE.
Установка Ullage
У этого сервера нет опубликованного пакета — он собирается из исходников. Открой репозиторий и следуй инструкции в README.
▸ github.com/nukk-pain/ullageFAQ
Ullage MCP бесплатный?
Да, Ullage MCP бесплатный — установка в пару кликов через Unyly без оплаты.
Нужен ли API-ключ для Ullage?
Нет, Ullage работает без API-ключей и переменных окружения.
Ullage — hosted или self-hosted?
Self-hosted: сервер запускается локально на твоей машине командой из раздела установки.
Как установить Ullage в Claude Desktop, Claude Code или Cursor?
Открой Ullage на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.
Похожие MCP
Fetch
Web content fetching and conversion for efficient LLM usage.
AWS KB Retrieval
Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.
автор: modelcontextprotocolSpring AI MCP Server
Provides auto-configuration for setting up an MCP server in Spring Boot applications.
llm-analysis-assistant
A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also
автор: xuzexin-hzCompare Ullage with
Не уверен что выбрать?
Найди свой стек за 60 секунд
Автор?
Embed-бейдж для README
Похожее
Все в категории ai
