skill-autobench

Warn

Audited by Snyk on Aug 13, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (medium risk: 0.30). The skill’s runtime workflow mines “Brain conversation archive” pages under conversations/ (populated by a prior archive skill) and scans per-harness session transcripts, which are outsider-authored free text submitted by end users in those chats/transcripts before generating the eval.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 13, 2026, 06:05 PM
Issues
1
Security Audit — snyk — skill-autobench