split-pr

Warn

Audited by Snyk on May 3, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly directs the agent to fetch PR title/body and diffs with "gh pr view" / "gh pr diff" and to parse the repository .github/CODEOWNERS in NVIDIA/Megatron-LM — all user-generated/public GitHub content that the agent reads and uses to decide and execute PR-splitting actions, so untrusted third-party content can influence behavior.

Issues (1)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
MEDIUM
Analyzed
May 3, 2026, 07:06 AM
Issues
1
Security Audit — snyk — split-pr