generating-character-and-background-images

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from local JSON files that is then used in prompts for external image generation.
  • Ingestion points: characters.json and backgrounds.json (specifically the prompt fields) processed in scripts/generate_images.py.
  • Boundary markers: Absent; prompt strings are interpolated directly into the API request payload.
  • Capability inventory: The script performs file writes to the project directory and network requests to third-party APIs.
  • Sanitization: Absent; the script does not validate or sanitize the prompt strings before submission to the image generator.
  • [DATA_EXFILTRATION]: The skill uploads generated images to imgbb.com, a public third-party hosting service. While this facilitates permanent hosting for the pipeline, it constitutes data exposure to an external domain not on the standard whitelist.
  • [EXTERNAL_DOWNLOADS]: The script fetches image assets from api.kie.ai during the generation process and saves them to the local characters/ and backgrounds/ directories.
  • [COMMAND_EXECUTION]: The skill requires the execution of a Python script that modifies local JSON files, creates environment configuration files, and communicates with external web services.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 02:15 AM
Security Audit — agent-trust-hub — generating-character-and-background-images