app-store-screenshots
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill includes a command to download and install the vendor's CLI tool using
curl -fsSL https://cli.inference.sh | sh. This is an installation script provided by the author of the skill (inference-sh-9) and is documented as a binary downloader with checksum verification. - [COMMAND_EXECUTION]: The skill requires the
Bashtool with permission to execute theinfshcommand. This tool is used to run various AI applications for image generation, video production, and image stitching as part of the screenshot creation workflow. - [PROMPT_INJECTION]: The skill includes a surface for indirect prompt injection (Category 8) as it interpolates user-provided text into JSON payloads for the
infshcommand. - Ingestion points: User-provided strings are passed to the
promptfield in JSON objects within bash commands (e.g., in SKILL.md). - Boundary markers: Commands use single-quoted JSON strings to encapsulate user input, which provides basic structural separation.
- Capability inventory: The skill uses the
Bashtool to execute theinfshCLI, which interacts with external AI services. - Sanitization: No explicit sanitization or filtering of the user-provided prompts is mentioned in the skill instructions.
Audit Metadata