ideogram-hello-world
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows security best practices by using environment variables for API keys and communicating with official, well-known service domains for the Ideogram API.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-defined prompts for image generation, which is its primary function. The vulnerability surface is mitigated by the external service's content filters.
- Ingestion points: The prompt parameter in the JSON and multipart form bodies within SKILL.md.
- Boundary markers: None.
- Capability inventory: Network operations via curl, fetch, and requests; file system write access via standard library functions (fs, open).
- Sanitization: Relies on Ideogram API content moderation and returns a safety status field (is_image_safe).
Audit Metadata