cross-engine-audit
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided novel projects (untrusted data) through multiple analysis engines. Malicious instructions embedded within the novel text could potentially influence the behavior of these engines or the content of the consolidated audit reports. * Ingestion points: The content of the novel project files under audit (e.g., scenes and chapters). * Boundary markers: No explicit delimiters or 'ignore instructions' warnings are provided to separate the data being analyzed from the agent's instructions. * Capability inventory: The skill coordinates multiple analysis engines and performs file writes to 'audit/AUDIT_RUN.md' and 'quality/FINDINGS.md'. * Sanitization: There are no mentioned mechanisms for sanitizing or escaping the content retrieved from the novel files before it is processed by the engines.
Audit Metadata