vectora

Name: vectora
Author: brunosrz

Local

github.com/brunosrzAI & Agents1Updated today

Repository

About vectora

Vectora is a self-hosted AI assistant app for people who want RAG, web search, filesystem access, and memory. This monorepo contains the app alongside its company site, documentation, and gateway and update services.

Key features

01Self-hosted AI assistant app
02RAG, web search, and memory
03Filesystem access
04Python backend with Vite and Electron
05Native installers and update channel

Use cases

01Run an AI assistant in a self-hosted setup
02Use web search and local files as context
03Build and distribute native app releases

MCP support

The tagline describes Vectora as an MCP sub-agent for Claude Code, but the README does not explain its MCP capabilities.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config

{
  "mcpServers": {
    "brunosrz-vectora": {
      "command": "uvx",
      "args": [
        "vectora-agent"
      ],
      "env": {
        "GOOGLE_API_KEY": "<GOOGLE_API_KEY>",
        "COHERE_API_KEY": "<COHERE_API_KEY>",
        "TAVILY_API_KEY": "<TAVILY_API_KEY>",
        "OPENAI_API_KEY": "<OPENAI_API_KEY>",
        "ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>",
        "LANGSMITH_API_KEY": "<LANGSMITH_API_KEY>"
      }
    }
  }
}

Run commanduvx vectora-agent

Environment variables to set

GOOGLE_API_KEYSecretGoogle Gemini API key (recommended free-tier LLM provider)

COHERE_API_KEYSecretCohere API key — required for RAG embeddings and reranking

TAVILY_API_KEYSecretTavily API key — required for web search and URL extraction

OPENAI_API_KEYSecretOpenAI API key (optional — alternative LLM provider)

ANTHROPIC_API_KEYSecretAnthropic API key (optional — alternative LLM provider)

LANGSMITH_API_KEYSecretLangSmith API key (optional — tracing and observability)

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages

pypivectora-agent0.1.0rc3stdio

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Vectora — Monorepo

Monorepo da Vectora. Cada produto vive em sua própria pasta, com toolchain e lockfile independentes. O que conecta tudo é o GitHub Actions (um workflow por projeto, gated por path) e o pre-commit compartilhado na raiz.

Estrutura

Pasta	Projeto	Stack	Deploy
`vectora/`	App Vectora (assistente de IA self-hosted)	Python (uv) + Vite + Electron	Instaladores nativos (Nuitka + Electron) via canal de update
`company/`	Site institucional	TanStack Start + Vite (pnpm)	Vercel → vectora.company
`docs/`	Documentação	Hugo	Vercel → docs.vectora.company
`services/`	Gateway (OAuth/webhooks pro desktop, ex-relay) + updates (distribuição de releases) — era `relay/` + `update-server/`, unificados	Hono + Cloudflare Workers (pnpm)	Cloudflare (wrangler)

O app em si (backend Python, frontend Vite, casca Electron) fica todo dentro de vectora/ — veja vectora/README.md para detalhes de arquitetura, build e CLI.

CI/CD (`.github/workflows/`)

4 workflows reais, cada um gated por path (on.push/pull_request.paths): uma mudança só no vectora/ não dispara o build/deploy de company/docs/services.

Workflow	Dispara em	Faz
`vectora.yml`	`vectora/*`, tags `v`	lint • security • build • testes (unit/integration/e2e) • release nativo (Nuitka + Electron) • publica no canal de update (R2 + KV)
`edge.yml`	`company/`, `services/`, `docs/**`	lint/typecheck/test dos 3 projetos • deploy automático em push em master (Vercel + Cloudflare)
`release-please.yml`	push em master	mantém o PR de release acumulado (Conventional Commits) — mesclar publica a tag `vX.Y.Z`
`pr-checks.yml`	pull requests	labels automáticos • valida título em Conventional Commits

Não existe workflow de Docker Build/Push, CodeQL nem triage de issues — a distribuição do app vectora é via instaladores nativos (não imagem Docker).

Secrets esperados

Secret	Usado por
`GOOGLE_API_KEY`, `COHERE_API_KEY`, `TAVILY_API_KEY`, `LLM_PROVIDER`, `LOG_LEVEL`	testes do vectora
`WIN_CERTIFICATE_BASE64`, `WIN_CERTIFICATE_PASSWORD`, `APPLE_ID`, `APPLE_APP_SPECIFIC_PASSWORD`, `APPLE_TEAM_ID`, `VECTORA_RELEASES_TOKEN`	assinatura/publicação dos instaladores nativos
`VERCEL_TOKEN`, `VERCEL_ORG_ID`, `VERCEL_PROJECT_ID_COMPANY`, `VERCEL_PROJECT_ID_DOCS`	deploy Vercel (company + docs)
`CLOUDFLARE_API_TOKEN`, `CLOUDFLARE_ACCOUNT_ID`, `R2_ACCESS_KEY_ID`, `R2_SECRET_ACCESS_KEY`	deploy do services + canal de update (R2)

Desenvolvimento

Cada projeto é independente — entre na pasta e use o gerenciador dele:

# App Vectora (Python + Vite + Electron)
cd vectora && uv sync && uv run vectora start

# Site institucional
cd company && pnpm install && pnpm dev

# Docs (Hugo)
cd docs && hugo server

# Services (gateway + updates)
cd services && pnpm install && pnpm dev

Pre-commit (raiz)

O .pre-commit-config.yaml na raiz cobre todos os projetos (ruff/ty/bandit no vectora/, prettier/oxlint/tsc no frontend, actionlint nos workflows):

pre-commit install
pre-commit run --all-files

Licença

Proprietária. Consulte LICENSE.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

Read the getting-started guide

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.