Meta’s Muse leaks internal files via prompt injection, developers say
Two developers independently extracted Muse’s entire filesystem—including Ubuntu system files, app templates, and internal documentation—using simple prompts. Peter James and Jonny L. Saunders found Muse generated zipped archives of its root directory with minimal resistance, calling its prompt injection defenses ‘almost nonexistent.’ Meta dismissed the findings as non-security-critical, noting…
Key points
- Developers extracted Muse’s full filesystem via prompts, bypassing security checks with minimal effort
- Leaked files include Ubuntu system files, internal docs, and scripts—some allegedly written by Claude
- Meta denies infrastructure risk but admits ongoing product changes may limit exposed data
Model page: Muse →
The story so far
3 episodes →- Meta’s Muse leaks internal files via prompt injection, developers saythis story
Muse will apparently let you download its entire filesystem
The Verge AI · 24 September 2026
Loading the full article…
This text was published by The Verge AI and written by Terrence O’Brien. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
2sources- Meta makes the Muse filesystem even more accessiblePress · The Verge AI ·
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Agents & Tools
All →- OpenAI launches Dots agent for ChatGPT Pro and Business Premium · 4 src
- Claude Code builds a file-organizing AI agent in 30 minutes · 1 src
- Anthropic lets users mod Claude Code like video game plugins · 2 src
- Cresta builds Conductor agent builder on Claude Agent SDK · 1 src
- GitHub launches ReviewBench to test AI code review agents · 3 src
Comments
via GitHub Discussions