Would you let an AI agent write to your production database?
Hi everyone, I'm Ibrahim, the solo developer behind Gadwal.
I built it because I let Claude Code work on my own Cloudflare D1 and Postgres databases. Reads were great. Writes made me nervous, because a plain database MCP connection runs whatever SQL the agent sends.
In Gadwal every connection has an AI level: Off, Read-only, Ask before write or Full. Production connections start at Read-only. On "Ask before write", an AI change waits in the table with its exact SQL until you approve, edit or reject it.
Gadwal launches here tomorrow (20 Nov). Before that, I'd like to hear how you handle this today:
Do you give AI coding tools access to real databases? Read-only, or writes too?
What would you need to see before you approved an AI-written UPDATE or DELETE?
Which database would you connect first? Gadwal 1.0 works with Postgres, MySQL, MariaDB, SQLite, DuckDB, Cloudflare D1, ClickHouse and BigQuery.
I'll read every reply and use them to plan the next update.

p/runevr
p/producthunt
p/google-cloud-platform
p/ai-context-flow
p/jot-2-0
p/beamfy
p/refs
p/dex-by-exmergo
p/reso
p/kodlo
p/duckfightclub
p/seshka
p/talanoa-2
p/nudgeforme
p/nexupira
p/dino-kids
p/many-displays
p/murror
p/operum
p/supabase
Replies
Be the first to reply
Have a question or a thought to share? Add a comment above to start the conversation.