Selvedge is a local MCP server that lets AI coding agents log structured change events and reasoning into a SQLite database under .selvedge/ next to a codebase. It also provides commands to query that history, including blame, prior attempts, diffs, search, and stale decisions, so users can recover why changes were made.
From the registry: Long-term memory for AI-coded codebases — agents log the why behind every change and read it back.

$ pip install selvedgehttps://pypi.org/project/selvedge/$ selvedge initPlease install the `selvedge` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Selvedge Python package** — Install the Selvedge package so the MCP server command is available on your PATH. Run: `pip install selvedge` (https://pypi.org/project/selvedge/)
- **Initialize Selvedge in your project** — Run once inside each project to create the local Selvedge database under .selvedge/. Run: `selvedge init`
Canonical MCP server config (stdio transport):
- command: `selvedge-server`
- args: []
- optional environment variables:
- `SELVEDGE_DB`: Optional path override for the Selvedge SQLite database. If unset, Selvedge uses the project database created by `selvedge init`, or falls back to the global database location. (example: `<path-to-selvedge-db>`)
Note: The README recommends `selvedge setup` as the easiest path because it installs MCP wiring and agent instructions automatically, but the canonical manual MCP server entry is a stdio server using the `selvedge-server` command with no args. Selvedge stores data locally in SQLite.
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.MCP server for interacting with the Supabase platform