eve-forge
Warn
Audited by Snyk on Jul 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.75). SKILL.md describes an “Absorb” stage where the orchestrator reads a business’s transcripts/examples and then the “Author” stage writes
agent/instructions.md/tool inputs from templates; those absorption inputs are outsider-authored free text (e.g., transcripts/examples not written by the operating user) that the agent would ingest into its LLM context to producetenant-spec.jsonand authored agent files.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata