media-planning
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data from various sources, which presents a risk of indirect prompt injection if those sources contain malicious instructions.
- Ingestion points:
SKILL.md(Workflow Step 8) and the "Evidence Produced" table specify the use of platform exports, station logs, and third-party audience audits as inputs for media planning and post-buy reconciliation. - Boundary markers: The instructions lack explicit boundary markers (e.g., XML tags or delimiters) or specific guidance to the agent to disregard any natural language instructions found within these external documents.
- Capability inventory: While the skill states capabilities are limited to "Read and search only" for analysis, it does not specify restrictions on the tools available to the agent for processing these files.
- Sanitization: There are no documented steps for validating, filtering, or sanitizing the content of external logs and reports before they are incorporated into the agent's context.
Audit Metadata