Stop letting your AIs live in silos — give them one shared database
WHAT IT SOLVES
You build a todo in Claude, rebuild it in Cursor, rebuild it again in ChatGPT. Three AIs, three memories, zero coordination
WHY IT'S INTERESTING
Every change is attributed — human or which agent
Most AI tools treat memory as a black box — you have no idea who changed what and when. Statey tags every mutation with the human or agent that made it, plus a versioned schema and full history. This isn't database flexing, it's solving a real coordination problem when multiple AIs work on the same stuff
MCP-native — one line to connect, zero infra to run
Instead of shipping another self-hosted open-source project, the author went hosted + native MCP. One command (`mcp add mcp.statey.ai`) and it works in Claude, Cursor, any MCP client. Lower barrier to entry, and it shows the author thought clearly: solo devs won't self-host a database just to share state between AIs
TECH GUESS
Hosted Postgres-ish DB service, likely Next.js frontend, exposes records via MCP protocol endpoints
Discussion (0)
- No comments yet — be the first.
Related
#229▶ 135Upload your textbook. AI teaches it back to you, chapter by chapter
#228▶ 111Someone needs to watch the AI spend meter
#227▶ 183Your users are kittens. Don't let them fall off the cliff