self-reflection
Warn
Audited by Snyk on Jul 31, 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 the self-reflection runtime, the skill reads outsider-authored user/assistant messages from locally stored session JSONL transcripts (tail -50 / tail -n 50) via
openclaw sessions --active ... --jsonand thentailon~/.openclaw/agents/main/sessions/<sessionId>.jsonl, which can include text originally submitted by external users (e.g., Telegram/direct chats).
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata