ad-performance-loop
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process metrics and creative data from external ad networks.
- Ingestion points: Performance metrics are gathered from Meta Ads Manager, TikTok Ads, Google Ads, and MMP/UTM-based attribution sources (SKILL.md).
- Boundary markers: Absent. The skill instructions do not specify delimiters or guidelines to distinguish untrusted external data from system instructions.
- Capability inventory: The skill utilizes
memory MCP,sequential MCP, andpostgres MCPfor storage and logic. There is no access to shell commands, arbitrary network requests, or file system modifications. - Sanitization: There is no evidence of sanitization or input validation for the ingested data.
Audit Metadata