Skills
skills/thepexcel/agent-skills/kie-ai/Gen Agent Trust Hub

kie-ai

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads generated media files from api.kie.ai and tempfile.redpandaai.co to the user's local directory.
  • [COMMAND_EXECUTION]: It provides a logic flow for generating and executing Python scripts to automate API calls and file management.
  • [DATA_EXFILTRATION]: Transmits user-provided prompts and media data to api.kie.ai and kieai.redpandaai.co for processing as part of the intended service functionality.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes external prompts for image generation.
  • Ingestion points: Prompt inputs provided to the createTask and veo/generate endpoints (file: SKILL.md).
  • Boundary markers: None specified in the API request templates.
  • Capability inventory: Network access to api.kie.ai and file system writes to D:/ClaudeMediaGen/output-fal/.
  • Sanitization: No explicit sanitization or validation of the input prompt is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 02:10 AM