Observed arrival · 2026-10-30
Agent Shakedown turns agent logs into a trust test
A forthcoming evaluation service for personal AI agents, with a configuration scan and a fuller shakedown that grades actions against logs and stated rules.
- For
- People deploying or expanding the permissions of personal AI agents
- Worth noticing
- Its published Otto report says a deliberately careless test agent failed all 7 safety gates; the site's own agent held them.
Field notes
The page separates a configuration scan from a graded live-style shakedown: the scan returns a short findings list by email, while the fuller test is described as checking defined errands, permissions, and confirmations against logs and outputs. Its first registry report covers 139 administrations and reports that Otto passed all seven listed safety gates, while a deliberately careless comparison agent failed all seven. The site also flags mismatches between Otto's completion claims and its logs.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue