imagegen
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill integrates untrusted user input and external data into its workflow, creating an attack surface where embedded instructions could potentially influence the agent's behavior.
- Ingestion points: Untrusted user prompts for image generation and user-provided image URLs or local file paths for image editing tasks (found in SKILL.md).
- Boundary markers: The skill instructions do not specify any delimiters or explicit warnings to the agent to ignore instructions that might be embedded within the image prompts or the content of the linked resources.
- Capability inventory: The skill performs network POST requests to a local API service (localhost:8402) and references the ability to read local file paths for image processing.
- Sanitization: There is no evidence of input validation, sanitization, or escaping of the user-provided prompts or file paths before they are sent to the image API.
Audit Metadata