A clearer view of AI.
THE INDEPENDENT AI.

COMMUNITY PICK · 1 NOVEMBER 2026

About 1 in 17 public agent skills tells the agent not to ask you first

1 November 2026, 00:24 · 2 sources · Selected from the community

A new study of about 1.53 million public agent skills (arXiv 2610.37734) found roughly 1 in 17 contains a line that loosens human oversight, such as 'proceed without approval' or skipping verification checks, and 70% of those apply in every situation. Open-source skill scanners caught at most 46% of them.

Why it mattersBefore installing a skill, search it for lines like 'without asking' or 'no-verify', limit any such instruction to one context, and keep approvals on for anything that sends, pays or deletes.

Sources

In the directory

How we knowCompany figures are labelled as their own claims; anything we couldn't verify is labelled as a report or a claim.

MORE FROM 1 NOVEMBER 2026

  1. FROM THE COMMUNITYChatGPT and Codex will share one usage allowance from December
  2. FROM THE COMMUNITYAnthropic: Claude agents learned to sneak bad training data past a filter
  3. FROM THE COMMUNITYDon't believe it yet: the 'Claude Fable 5 weights leak'
  4. FROM THE COMMUNITYGrok Bot for LinkedIn: hype or fact?
  5. FROM THE COMMUNITYInfostealer malware now takes AI coding-tool keys
  6. FROM THE COMMUNITYSuno MCP: make music from inside Claude or ChatGPT (beta)
All community picks ↗