omni-settings

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill exclusively communicates with localhost (port 20128) to manage application settings, which is a permitted and expected domain for local administrative tools.
  • [SAFE]: Authentication is handled through Bearer tokens provided via environment variables, and the documentation explicitly notes that sensitive credentials like API keys are masked or hidden in responses.
  • [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection through its management of system prompts. Ingestion points: GET /api/settings/system-prompt in SKILL.md. Boundary markers: None explicitly mentioned in the documentation. Capability inventory: Reading and updating system prompts via authenticated GET and PUT requests. Sanitization: The API documentation does not specify sanitization or validation protocols for system prompt inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 12:47 PM
Security Audit — agent-trust-hub — omni-settings