observability
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of helper functions and component templates for application health monitoring. All functionality is aligned with the stated purpose of improving production visibility.
- [DATA_EXFILTRATION]: The skill implements reporting mechanisms that send error data, performance metrics, and console logs to internal API endpoints. This behavior is standard for observability tools and does not target sensitive user files or credentials.
- [PROMPT_INJECTION]: The user feedback component creates a surface for indirect prompt injection by transmitting unsanitized user input to the backend. Ingestion points: FeedbackReporter textarea in SKILL.md; Boundary markers: Absent; Capability inventory: fetch network operations in SKILL.md; Sanitization: None.
Audit Metadata