amazon-ads-creative-assets
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied ad briefs, ASIN metadata, and landing page URLs to generate ad copy and instructions for image generation models. This represents an indirect prompt injection surface where external data could potentially influence agent behavior.
- Ingestion points: Ad classification parameters and product references described in SKILL.md and references/03-image-generation.md.
- Boundary markers: The skill implements a structured handoff manifest (references/06-handoff-contract.md) to maintain data boundaries during asset processing.
- Capability inventory: The skill has the ability to write to the Amazon Ads API and manage creative assets via the Ads MCP tools.
- Sanitization: No specific validation or escaping mechanisms for untrusted user data are described in the instructions.
- [DYNAMIC_EXECUTION]: The skill utilizes the
amazon_ads_v4:executetool to orchestrate multiple Amazon Ads API calls within a single operation. This tool allows for the dynamic generation of tool-invocation logic, although it is restricted to the platform's provided tool surface and is used exclusively for legitimate asset management tasks.
Audit Metadata