Daily log

Today in agent chaos

Short daily notes on AI agents misbehaving. Items tagged As reported come from public reports we haven’t independently confirmed. The big ones get a full write-up in the Hall of Shame.

22 Oct 2026

A developer says Codex, cleaning up one temp folder, fell back to a badly quoted rmdir command and deleted nearly everything on their 2 TB D: drive. As reported
A developer says Claude Code, asked for a small doc-scraping workflow, built an eval harness nobody asked for, launched 44+ sessions on their login and read files far outside the project. As reported
A developer says a Claude Code subagent left a -1 placeholder in a SIGKILL cleanup line, then ran it, killing every process they owned: tmux, other Claude sessions, the desktop. As reported

21 Oct 2026

A developer says Claude Code symlinked a real repo into a test sandbox, then ran a script whose rm -rf deleted the repo's .git folder, with no remote or backup. As reported
A developer says Claude Code scripted a find-and-replace that ran past the function it was editing and silently deleted 17 unrelated tests. As reported

20 Oct 2026

Hidden text on about 80 WhatsApp help pages told Meta's support chatbot not to recommend contacting a data protection regulator. Meta blamed a bug and pulled it. Verified
SaaStr says a well-known vendor's AI agent deleted almost all its Salesforce deals within minutes of being connected. A same-day snapshot saved them. As reported
A developer says Claude Code, set up never to run rm, got blocked twice and then switched to find -delete to remove the files anyway. As reported

19 Oct 2026

OpenCode says a break-in at a legacy system may have exposed AI provider keys saved by 1.4% of its Console users. Some users say they've already been charged for usage they didn't run. Verified
A developer says Claude Opus 5.5 traced their $90,000 Google bill to Codex publishing their .env file on Cloudflare Pages. As reported
A Codex user says a gpt-6-luna agent emptied their whole project folder when a venv --clear command ran with a blank path. As reported

18 Oct 2026

OpenAI discloses four more agent incidents from June, including a model that pulled credentials from a government statistics site and one that downloaded 147,451 rows of non-public data. Verified
JPMorgan and Charles Schwab say they're blocking outside AI agents from getting into customer accounts and moving money. Verified
More Cursor users on Windows report the agent running a delete command that wiped a whole drive instead of one folder. As reported