skills/wubin1836/ai-hive-agent-skills/ai-model-expert-viral-cover-image-generation-editing/Gen Agent Trust Hub
ai-model-expert-viral-cover-image-generation-editing
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads generated media files such as images and videos from the vendor's domain (ai-hive.iclip.cn) to the local Downloads directory.
- [COMMAND_EXECUTION]: Instructions guide the agent to run a local Python script (scripts/imagegen.py) to interface with the AI-HIVE API.
- [DATA_EXFILTRATION]: User-provided text prompts and local media files are uploaded to the vendor's API endpoint (ai-hive.iclip.cn) for processing, which is the intended core functionality.
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection through its ingestion of external content.
- Ingestion points: Data enters via the --prompt and --image arguments of the imagegen.py script.
- Boundary markers: The script does not use explicit boundary markers or delimiters for the ingested content.
- Capability inventory: The tool has network access to the vendor's API and file-write access for downloading results.
- Sanitization: No sanitization or filtering is performed on the input before transmission to the generative model.
Audit Metadata