campaign-challenger
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by ingesting outbound sales sequence drafts and historical campaign data to perform quality audits and benchmarking.
- Ingestion points: The skill retrieves sequence drafts and historical campaign performance data (including message copy) as specified in references/benchmark-method.md.
- Boundary markers: No explicit delimiters or boundary markers are defined for the content being analyzed, increasing the risk of the agent misinterpreting message copy as instructions.
- Capability inventory: The skill is designed to analyze, score, and propose fixes for campaign copy, with instructions in SKILL.md and references/quality-check.md suggesting the agent can write or modify sequence files.
- Sanitization: There is no evidence of sanitization or escaping logic applied to the external campaign data before processing.
Audit Metadata