facebook-marketplace-safety-guard
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a structured security guard using a predefined regex catalog. It is a pure function that does not perform dynamic code execution, remote script downloads, or network exfiltration. The logic is entirely self-contained and auditable within the file.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted natural language messages from Facebook Marketplace, which is an attack surface for indirect prompt injection. • Ingestion points: Untrusted text enters the skill through the
textinput parameter. • Boundary markers: No explicit delimiters are used to wrap the untrusted message content. • Capability inventory: The skill metadata indicates it requires access to theterminaltoolset. • Sanitization: The input is processed directly by regex matches without escaping or validation, although the skill's purpose is to filter the data rather than execute it.
Audit Metadata