asc-app-shots-prompt
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data in the form of app screenshots provided by users. This constitutes a standard data ingestion surface common to vision-based tasks.
- Ingestion points: User-provided screenshot image files located at a user-specified path (SKILL.md, Step 1).
- Boundary markers: The skill does not define explicit delimiters for the image analysis output, but it enforces a strict 6-component structure for the generated prompt (SKILL.md, Step 3).
- Capability inventory: The skill itself does not execute commands; it generates text for the user to use with the
ascCLI tool. It lacks internal capabilities for network access, file writing, or automated subprocess execution. - Sanitization: The transformation of visual data into a descriptive text prompt by the AI model acts as a natural language filter, reducing the risk of direct instruction passthrough.
Audit Metadata