campaign-strategy
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute specific Python scripts located at
../../scripts/dynamic_router.py,../../scripts/skill_router.py, and../../scripts/neural_router.py. These scripts are used to manage complex request routing and build dependency graphs within the workspace. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data such as marketing objectives and evidence, which could influence the behavior of the internal routing scripts. While the skill includes a discipline to classify evidence, the interpolation of user data into command arguments represents a baseline injection surface.
- Ingestion points: User-supplied business objectives, audience insights, and brand proof as described in SKILL.md and references/skill-spec.json.
- Boundary markers: Instructions to classify material claims as fact, inference, assumption, or unknown.
- Capability inventory: Execution of workspace-local Python scripts via subprocess or shell tools.
- Sanitization: No specific input validation or sanitization routines for user data before script invocation are documented.
Audit Metadata