metrics
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata from the RevOS semantic model (including metric titles and descriptions) which serves as an ingestion point for potentially malicious instructions if an attacker can manipulate those fields.
- Ingestion points: The skill fetches data using the
api_readtool from resources such ascubes,tables, anddashboard-cardsas specified inSKILL.md. - Boundary markers: The instructions do not define delimiters or provide specific warnings to the agent to ignore or isolate instructions that may be embedded within the retrieved metadata.
- Capability inventory: The skill relies on the
api_readcapability to gather and present data to the user. - Sanitization: There is no evidence of sanitization, validation, or filtering of the retrieved metadata before it is used to generate the agent's response.
Audit Metadata