classifier-lab-subagent
Warn
Audited by Snyk on Aug 26, 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). In orchestrator.py, _poll_logs reads progress.jsonl/result.json written by the scillm-driven subagent, so outsider-authored text can influence JSON fields the LLM-ingesting workflow later parses and logs (though it is “agent output” rather than direct outsider input).
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata