An MCP server for markdown vaults that identifies dormant notes using wikilink graph structure and file modification time, then surfaces them for rediscovery. It also finds orphaned notes, cross-domain bridges between notes, and can run a replay cycle that samples dormant notes with bias.
From the registry: Surface the notes you forgot you wrote — DMN replay for markdown vaults.
$ uvx mcp-curiosity-enginePlease install the `mcp-curiosity-engine` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Python 3.10+** — Python 3.10 or newer is required to run mcp-curiosity-engine.
- **Install the package runner or package** — Use the portable runner path with uvx, or install the package with pip. Prefer uvx for the MCP config command. Run: `uvx mcp-curiosity-engine`
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["mcp-curiosity-engine"]
- optional environment variables:
- `CURIOSITY_ENGINE_LICENSE`: Pro tier license key to unlock paid tools. Not required for free tools. (example: `<your-license-key>`)
Note: The README shows both pip install and uvx usage. The simplest portable MCP config is to run the server with uvx. A Pro license can also be supplied via CLI flag (--license-key <your-license-key>) or a local config file, but environment variable form is the canonical config representation.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.CURIOSITY_ENGINE_LICENSErequiredLicense key used to unlock the Pro tier features.Comprehensive PostgreSQL documentation and best practices, including ecosystem tools