wiki-room
Warn
Audited by Snyk on Jul 20, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.65). SKILL.md:13-21 describes opening a subagent “room” whose prompt is constructed from the user’s question plus a list of wiki page IDs and then (per Step 1) retrieves episodic content via
python3 harness/scripts/mem-rank.py retrieve "<câu hỏi>", which can cause outsider-authored free text (e.g., wiki page contents or retrieved episodes not authored by the operating user) to be ingested into the room’s LLM context at runtime.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata