campaign-report
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data ingested from external outbound platforms (PlusVibe, Instantly) via Model Context Protocol (MCP) tools. While primarily for reporting, malicious content within campaign fields could attempt to manipulate the output or behavior.
- Ingestion points: Campaign metrics are pulled via MCP during the first step of the reporting process described in SKILL.md.
- Boundary markers: The instructions lack specific delimiters (e.g., XML tags) or instructions to ignore embedded commands within the pulled metrics.
- Capability inventory: The skill is scoped to data synthesis and formatting; it does not request or use capabilities for shell execution, file system modification, or non-whitelisted network requests within its logic.
- Sanitization: There is no evidence of sanitization or schema validation for the data retrieved from the external campaign platforms.
Audit Metadata