Thursday Jun 18
COOL TOOLSOther

Simon Willison Ships A Datasette Agent

18JUN
APPROVE TO WRITEASKS FIRSTWRITES

Your database now has an AI assistant that asks before it writes. Simon Willison shipped Datasette Agent, an LLM that runs and edits SQL with your approval. A small answer to a big agent-safety problem.

Datasette is Willison's open-source tool for exploring data. The new agent reads your tables, writes queries, and can edit rows.

The interesting part is the brake. A new tool, execute_write_sql, stops and asks before changing data. The human stays in the loop on every write.

Most agent demos let the model touch production and hope. This flips it. Approval is the default, not the afterthought.

full brief & sources

⚡ Why this matters

  • A named, trusted builder shipping a concrete pattern for safe agent database access.
  • Approval-gated writes are the missing piece in most agent-touches-your-data demos.
  • Open source, so the pattern spreads to every tool that copies it.

🔍 What happened

  • Datasette Agent is an extensible LLM assistant built into Datasette.
  • It can read schema, run read queries, and write to the database.
  • A new execute_write_sql tool asks for user approval before any write.
  • Flags include --yes to auto-approve and --unsafe to skip the brakes.
  • Shipped in mid-June 2026 as part of the datasette-agent package.

💬 Smart takes

  • Simon Willison: the agent asks user approval, then writes to the database, taking user permissions into account.
  • Skeptic: an approval prompt only helps if the human reads it, and click-through fatigue makes approve-all the real default.

🧭 Where this goes

  1. Likelyapproval-gated tool calls become a standard pattern in agent frameworks this year.
  2. Likelyenterprise data tools copy the read-versus-write permission split for agents.
  3. Possiblean --unsafe agent incident makes the news and proves the brake's value.
  4. Wild CardDatasette Agent becomes the reference design that bigger vendors quietly clone.

🥄 The Spoon Take

The agent hype is all about what the model can do. The quiet, important work is about what it should be allowed to do. Willison's answer is boring and right: let the agent propose, make the human approve, log everything. That's the pattern production teams will actually ship.

🤔 Pushback

One open-source tool from one builder is not an industry standard, and most teams will skip the approval step the moment it slows them down.

·Datasette - Agent announcement·GitHub - datasette-agent·Simon Willison - Datasette tag