ads-monitor
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data in the form of 'normalized snapshots', which creates a potential surface for indirect prompt injection if the data contains malicious instructions. 1. Ingestion points: Instructions in SKILL.md specify loading two or more account snapshots. 2. Boundary markers: There are no specific delimiters or instructions to ignore commands within the snapshots. 3. Capability inventory: The skill is restricted to analysis and persistence of results, with an explicit instruction not to mutate the account. 4. Sanitization: No content validation or sanitization of the snapshot data is mentioned.
Audit Metadata