product-led-growth
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions or scripts.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data via prompt placeholders in SKILL.md (e.g., product descriptions) without explicit boundary markers or sanitization. Ingestion points: User-supplied placeholders such as [description] and [audience] in the SKILL.md prompt examples. Boundary markers: Absent. Capability inventory: The skill includes scripts/main.py which performs basic arithmetic calculations; it has no network or file system capabilities. Sanitization: Absent. This poses no significant security risk due to the lack of exploitable capabilities.\n- [METADATA_POISONING]: A minor inconsistency exists between the author handle in the metadata ('ClawFu') and the provided author context ('guia-matthieu'). This is documented for completeness but does not indicate malicious intent.
Audit Metadata