PromptArmor company blog
Claude Cowork exfiltrates files
Claude Cowork is vulnerable to file exfiltration attacks via indirect prompt injection as a result of known-but-unresolved isolation flaws in Claude's code execution environment.
- Points
- 870
- HN comments
- 399
- Submissions
- 1
- Cited by
- 4 commenters
AI & Machine LearningSecurity & Privacy2026
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Jan 14, 2026 | Claude Cowork exfiltrates files by takira | 870 | 399 |
Cited in comments
Linked from 4 Hacker News comments by 4 different people, from Jan 2026 to Aug 2026.
Most recent:
- htrp · Aug 5, 2026
- jingkai_he · Feb 8, 2026
- yunwal · Jan 21, 2026
- rvz · Jan 20, 2026
