generate-image
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill makes network requests to
openrouter.ai, which is a well-known service for accessing AI models. This interaction is necessary for the skill's core functionality and does not involve downloading executable code from untrusted sources. - [PROMPT_INJECTION]: The skill ingests natural language prompts to generate or edit images, creating a surface for indirect prompt injection.
- Ingestion points: The
promptargument inscripts/generate_image.py. - Boundary markers: None; the prompt is passed directly into the API request payload.
- Capability inventory: Filesystem read (input image), filesystem write (generated PNG), and network egress (API call to OpenRouter).
- Sanitization: None; the script does not escape or validate the text prompt before transmission.
- [CREDENTIALS_UNSAFE]: The script implements a safe method for handling API credentials by searching for an
OPENROUTER_API_KEYin a local.envfile or environment variables, avoiding the use of hardcoded secrets.
Audit Metadata