qinghu-scene-image
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download the Node.js runtime from its official website and install the @iqinghu/qhkit package from the npm registry. These downloads target official, well-known infrastructure.\n- [COMMAND_EXECUTION]: The skill relies on shell commands and the qhkit CLI tool to perform its primary functions and set up the execution environment.\n- [PERSISTENCE]: To maintain the environment configuration across sessions, the instructions suggest appending the Node.js binary path to the user's shell profile file.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text for product names and scene descriptions, which are then passed to the image generation tool.\n
- Ingestion points: User input is incorporated into the prompt field within the JSON payload for the qhkit CLI command in SKILL.md.\n
- Boundary markers: The skill utilizes specific delimiters in its command templates to separate user content from system instructions.\n
- Capability inventory: The skill possesses the ability to execute shell commands, perform network operations via qhkit, and read local image files.\n
- Sanitization: The skill does not explicitly perform sanitization or validation on the user-provided strings before they are processed by the underlying tool.
Audit Metadata