revid-product-description-to-ad
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The
examples/run.shscript performs automated testing by callingcurlto interact with the Revid API andjqto process responses. The script safely incorporates user data into the request payload usingjqarguments to prevent shell-level injection. - [INDIRECT_PROMPT_INJECTION]: The skill transforms untrusted product descriptions into advertising scripts through the Revid service.
- Ingestion points: Product description text provided in the
promptfield and optional brand notes instylePrompt. - Boundary markers: Not present; the instructions treat the input as a creative brief.
- Capability inventory: Execution of the Revid API workflow and associated shell commands in the example test script.
- Sanitization: None; the skill relies on the integrated service provider's internal safety measures for content generation.
Audit Metadata