live-dashboard
Pass
Audited by Gen Agent Trust Hub on May 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill generates executable HTML and JavaScript code but mitigates risks by including a dedicated sanitization function
e()to escape HTML special characters before rendering dynamic data, protecting against cross-site scripting (XSS) and indirect prompt injection. - [SAFE]: Authentication tokens and integration secrets are handled via a local daemon proxy (
/api/od/connectors/poll) and resolved server-side, preventing credential exposure within the generated artifact. - [SAFE]: The skill uses the
file_writecapability appropriately to create dashboard and configuration files within the project workspace, with no evidence of unauthorized file system access or persistence mechanisms. - [SAFE]: No obfuscation, metadata poisoning, or unauthorized privilege escalation patterns were detected in the skill's instructions or templates.
Audit Metadata