benchmark-to-brief
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data that could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill is designed to ingest "benchmark reports," "master tables," and "comment analyses" (SKILL.md).
- Boundary markers: The skill includes a "Fact Rule" and "Judgment Flow" to distinguish between observed facts and inferences, which serves as a prompt-level constraint (SKILL.md).
- Capability inventory: The skill documentation references the execution of "owned CLI or script commands" as part of its operational boundary (SKILL.md).
- Sanitization: There are no explicit instructions or mechanisms mentioned for sanitizing or escaping the content of ingested research artifacts before they are used to determine which CLI commands to execute.
Audit Metadata