creativeclaw-generate-image

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a standard operational workflow for generating images, including model selection and prompt construction guidelines tailored to specific image models.- [EXTERNAL_DOWNLOADS]: The configuration in agents/openai.yaml specifies an MCP tool dependency at https://app.creativeclaw.co/mcp/chatgpt. This endpoint is owned by the skill's vendor and is used for secure tool communication.- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from search_assets, get_example, and get_theme (SKILL.md) to build prompts for image generation. No explicit boundary markers or sanitization logic is described to isolate retrieved asset metadata from model instructions. The skill utilizes generate_image, list_models, and check_job capabilities (SKILL.md). This represents a functional surface for indirect prompt injection but aligns with the skill's intended purpose of creative asset retrieval.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 08:24 PM
Security Audit — agent-trust-hub — creativeclaw-generate-image