02-platform-audit
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines a standard business workflow for social media platform audits, focusing on manual review of public profiles and official tools like the Meta Ad Library. No dangerous execution patterns, privilege escalation attempts, or persistence mechanisms are present in the provided instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface by ingesting external data such as client handles and competitor names. However, the risk is negligible because the skill lacks the capabilities to perform dangerous operations like shell command execution or network exfiltration.
- Ingestion points: Data enters the skill via required inputs such as client handles, platform URLs, and the previously completed client brief (SKILL.md).
- Boundary markers: The skill does not implement specific delimiters or 'ignore instructions' warnings for the processing of external data.
- Capability inventory: The skill is restricted to generating text and markdown tables; it contains no subprocess calls, remote code execution patterns, or sensitive file system access.
- Sanitization: External content is interpolated directly into the audit reports and slide outlines without explicit sanitization or filtering.
Audit Metadata