convez.ai
MCP ServerIntegração & APIsIntermediário
DoubleGremlin181

Potluck — Base de Conhecimento Pessoal Local

DoubleGremlin181/potluck

Base de conhecimento privada local para potencializar seu AI com seus dados pessoais

Ver no GitHub
InicialPluginScore · de 0 a 100

MCP server que transforma seus dados pessoais (Google Takeout, GDPR exports) em uma base de conhecimento local, criptografada e sem nuvem. Expõe acesso via web app, CLI e MCP para integração com assistentes de IA, permitindo personalização profunda mantendo privacidade total.

Instalação

$npx convez-ai DoubleGremlin181/potluck

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
0
Forks
0
Acompanhando
0
Saúde
42%
Último commit
há 4 meses
Versão
—
Issues abertas
46
Contribuidores
1

Casos de uso

  • Integrar Google Keep e Gmail com Claude para busca contextual personalizada
  • Criar assistentes de IA treinados em seus próprios dados sem exposição à nuvem
  • Migrar dados de múltiplas fontes (GDPR exports) para um repositório unificado e pesquisável

Veredito

Projeto ambicioso em reescrita ativa (v1 alpha). Para quem prioriza privacidade e quer controle total dos dados pessoais conectados à IA — ainda em fase early, mas arquitetura sólida. Ideal para builders que querem base de conhecimento local com MCP nativo.

README

A tradução para o português deste README ainda está na fila.

Potluck

Privacy-first personal knowledge database for your AI — local-first, MCP-native.

Potluck ingests your data exports (Google Takeout first: Keep + Gmail; more sources each phase), stores everything in one local SQLite database, and exposes it three ways — a web app, a CLI, and MCP for AI assistants. No cloud, no telemetry: nothing leaves your machine.

Status — v1 rewrite in progress

PhaseShippable incrementStatus
P0 Reset & Walking Skeletonpotluck serve zero-config: SPA shell; CLI/API/MCP answer stats from one service layer; bench rig + CI✅ v1.0.0-alpha.1
P1 Storage Core & First IngestGoogle Keep from a Takeout zip; FTS search via CLI + MCP—
P2 Gmail at Scale & Search v1Multi-GB mbox ingested incrementally; filtered/snippeted search—
P3 MVP InterfacesReal search/item/imports UI, MCP toolset v1 — beta.1 = MVP—
P4 Source Expansion & AutomationAll v0 sources, watch-folder, scheduled GDrive pull—
P5 Semantic SearchUnified embedding space, HNSW index, hybrid RRF—
P6 Vision & Media EnrichmentOCR, image embeddings, faces, media gallery—
P7 Linkers, People & TimelineRelated-items everywhere, people review, timeline—
P8 Hardening & 1.0Docs, backup/restore, doctor, perf sweep—

Full plan, architecture, and locked decisions: pinned issue #98. v0 is archived at archive/v0.

Quickstart

Requires uv (Python is fetched automatically).

Run straight from GitHub — CLI, API, and MCP work; the web UI needs a built SPA, which this form does not include:

uvx --from git+https://github.com/DoubleGremlin181/potluck potluck serve

Release wheels ship with the web app embedded:

uvx --from https://github.com/DoubleGremlin181/potluck/releases/download/v1.0.0-alpha.1/potluck-1.0.0a1-py3-none-any.whl potluck serve

Or Docker (data persists in the potluck-data volume):

docker run -p 127.0.0.1:8765:8765 -v potluck-data:/data ghcr.io/doublegremlin181/potluck:1.0.0-alpha.1

Then open http://127.0.0.1:8765.

[!WARNING] Potluck is localhost-only by design — there is no authentication in v1. Do not expose it to other machines or the internet.

MCP for AI assistants

stdio (Claude Desktop, Claude Code, …):

{
  "mcpServers": {
    "potluck": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/DoubleGremlin181/potluck", "potluck", "mcp"]
    }
  }
}

Streamable HTTP instead: potluck mcp --http (default 127.0.0.1:8766).

P0 toolset: get_stats. Search and item tools land with P1.

CLI

potluck status      database overview (counts, paths, versions)
potluck serve       web app + API on one port (opens your browser)
potluck mcp         MCP server (stdio; --http for streamable HTTP)
potluck bench run   benchmark harness (smoke/full tiers)

Configuration is optional: defaults work out of the box. Override via POTLUCK_* env vars or ~/.config/potluck/config.toml (env > toml > defaults); the database lives at ~/.local/share/potluck/potluck.db by default.

Development

git clone https://github.com/DoubleGremlin181/potluck && cd potluck
uv sync && uv run pre-commit install
(cd web && npm ci && npm run build)
uv run potluck serve

Tests: uv run pytest (unit tier), -m browser for the Playwright smoke. Conventions live in CLAUDE.md; test patterns in tests/README.md.

License

MIT

Ver o original no GitHub

Bom para

Relacionados