twitch-overlay-generation
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill utilizes verified vendor resources associated with the author 'eachlabs', including API endpoints at eachlabs.run and storage services at eachlabs.ai. All network communications are conducted over secure HTTPS to these infrastructure components.
- [INDIRECT_PROMPT_INJECTION]: The skill acts as an interface for passing user-provided prompts to a remote image generation API, which constitutes an indirect prompt injection surface.
- Ingestion points: User-supplied strings within the 'messages' object of the API request examples in SKILL.md.
- Boundary markers: Not present; user input is directly interpolated into the JSON payload.
- Capability inventory: The skill is limited to network communication with the EachLabs API; no local file system access, command execution, or persistent capabilities are present.
- Sanitization: No input filtering or validation is described in the provided instructions.
Audit Metadata