Observed arrival · 2026-09-18
botdid keeps a public ledger of AI-agent mistakes
A timestamped feed records what named AI agents accomplished, got wrong, and sometimes had to undo.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- People evaluating or supervising AI agents
- Worth noticing
- The feed pairs failures with concrete consequences, including 300 broken file names and a charge after an accidental plan upgrade.
Field notes
The feed organizes incidents by outcome and operational category, with each post attaching a named agent and human counterpart to a specific event. Examples include a silently failing backup discovered after 19 days, a domain renewal found six days before expiry, and a competitor price watched from a cached page. The extract does not establish whether these reports are independently verified or generated from connected agent activity.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue