ContextVault

Shared context storage for AI agents. Any MCP-compatible model (Claude, ChatGPT, Gemini) can store, read, and search context entries through the MCP server. A web UI lets you browse and manage everything.

Architecture

  Claude / ChatGPT / Gemini
        │ (MCP stdio)
        ▼
  ┌──────────────┐
  │  MCP Server   │──────┐
  └──────────────┘      │
                         ▼
  ┌──────────────┐   ┌──────────┐
  │   Web UI      │──▶│ Express  │──▶ PostgreSQL
  └──────────────┘   └──────────┘

• Express API (src/index.ts) — REST API + serves the React UI
• MCP Server (src/mcp.ts) — stdio MCP server for AI agents
• PostgreSQL — persistent storage with JSONB tags and GIN index

Quick Start (Docker)

docker compose up

Open http://localhost:3000 for the UI. Postgres runs on port 5432.

Local Development

Prerequisites: Node.js 20+, PostgreSQL running locally.

# Create the database
createdb context_vault

# Install and build
npm install
cd client && npm install && npm run build && cd ..
npm run build:server

# Start the web server
npm start

# Or run the MCP server for AI agents
npm run mcp

Set DATABASE_URL env var if your Postgres isn't at the default postgresql://postgres:postgres@localhost:5432/context_vault.

MCP Configuration

Claude Code / Claude Desktop

{
  "mcpServers": {
    "context-vault": {
      "command": "node",
      "args": ["/absolute/path/to/context-vault/dist/mcp.js"],
      "env": {
        "DATABASE_URL": "postgresql://postgres:postgres@localhost:5432/context_vault"
      }
    }
  }
}

Deploy to GCP (Cloud Run)

# Build and push
gcloud builds submit --tag gcr.io/PROJECT_ID/context-vault

# Deploy
gcloud run deploy context-vault \
  --image gcr.io/PROJECT_ID/context-vault \
  --set-env-vars DATABASE_URL=postgresql://... \
  --port 8080 \
  --allow-unauthenticated

API

MCP Tools