- ID
- AIA-012
- Category
- AI AI agent incident
- Disclosed
- 2025-08
- Systems
- GitHub Copilot (VS Code agent mode)
- Harm type
- Prompt injection / code exec
- Status
- fixed
Rehberger writes: "With the August Patch Tuesday release this is now fixed." - CVE
CVE-2025-53773- CISA KEV
- Not in CISA KEV as of catalog 2026.10.30
Checked 2026-10-31 against catalog 2026.10.30
Summary
Johann Rehberger reported that a prompt injection could make GitHub Copilot write "chat.tools.autoApprove": true into VS Code settings ("YOLO mode") and then run commands without approval. He reported it on June 29, 2025, and he says it was fixed in Microsoft's August 2025 Patch Tuesday.
Controls
Controls that could prevent, detect, or support recovery from this kind of failure. Mappings are editorial judgments made by the method, not findings about what any party should have done.
NIST AI RMF
- MEASURE 2.7 Security and resilience are evaluated and documented
- MANAGE 2.4 Mechanisms exist to supersede, disengage, or deactivate the system
NIST SP 800-53
Human-approval (JCR) checkpoint
partial: approval gate the agent could disable by editing a settings file
Sources
- embracethered.com primary link checked 2026-10-31, HTTP 200
Post dated Aug 12, 2025
https://embracethered.com/blog/posts/2025/github-copilot-remote-code-execution-via-prompt-injection/
Corrections
No corrections recorded.
To report an error in this entry, see corrections.