Launching today
Trifecta Checker
A friendly pre-flight safety check for AI agents
2 followers
A friendly pre-flight safety check for AI agents
2 followers
AI agents now read mail, consume untrusted content, and send messages for us — Simon Willison's *lethal trifecta*: one poisoned message can steal private data. Trifecta Checker is a chat with Trafica, a friendly checker. Describe your agent or drop in an MCP config; Muse Spark builds a capability profile, a deterministic rule engine (never the LLM) rates it Low / Risky / Lethal, and Trafica explains it plainly, with gentle fixes. Six languages. Built for Meta's Global AI Developer Hackathon.
Free
Launch Team
Framer AI AgentsDesign and publish professional sites with AI
Promoted
Hi everyone — I'm Kentaro, a cloud architect from Japan. I built Trifecta Checker because the people handing powers to AI agents are usually not security specialists, and they deserve a check that speaks plain language and takes thirty seconds, *before* the agent is wired up.
The design rule I cared about most: the LLM never decides the verdict. Muse Spark translates in both directions (your words → capability profile, verdict → plain words), but a deterministic rule engine owns the rating, so the same setup always gets the same answer.
I'd love feedback: try describing your own agent setup, and tell me whether the verdict — and Trafica's questions back — felt right. Thank you!