cloak
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting and analyzing potentially untrusted external data, including codebases, log files, and database schemas, to detect PII. This creates a surface for indirect prompt injection where malicious instructions embedded in the analyzed data could attempt to influence the agent's behavior.
- Ingestion points: SKILL.md specifies scanning for PII in code, configs, logs, and database schemas.
- Boundary markers: The skill provides explicit instructions to "Sanitize before ingestion" and "use redacted examples only" to prevent accidental obedience to embedded data.
- Capability inventory: The skill is designed for analysis and generating code patterns; it does not explicitly request dangerous tools in its frontmatter, though it operates within the agent's provided environment.
- Sanitization: The skill includes robust guidance in reference/implementation-patterns.md for PII redaction and privacy-safe error handling to mitigate data leakage and exposure risks.
Audit Metadata