ads-performance-analytics
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to interpret data from external sources, specifically paid media dashboards and platform reports from Meta, Google, LinkedIn, and TikTok. This creates a potential vulnerability where malicious strings in campaign names or ad metadata could attempt to override the agent's instructions.
- Ingestion points: Paid media dashboards, platform-reported metrics, and warehouse revenue data (
SKILL.md). - Boundary markers: The skill does not provide specific instructions to delimit external data or warnings to ignore instructions embedded within the processed metrics.
- Capability inventory: The instructions refer to the use of platform-specific MCP commands for retrieving dashboard data, although no specific tools or scripts are included in this skill package (
SKILL.md). - Sanitization: There is no evidence of input validation, filtering, or sanitization requirements for the data ingested from external platforms.
Audit Metadata