add-announcement

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill provides a surface for indirect prompt injection by ingesting user-provided data to modify a configuration file that controls UI elements.
  • Ingestion points: The skill uses the Edit tool to modify web/oss/src/components/SidebarBanners/data/changelog.json with user-provided titles and descriptions.
  • Boundary markers: The skill documentation includes a mitigation note: 'Don't add custom HTML/styling in description
  • it's rendered as plain text.'
  • Capability inventory: The skill is authorized to use Read, Edit, Grep, and Glob tools to perform file system operations.
  • Sanitization: The skill relies on manual adherence to the 'plain text' instruction and the underlying UI's rendering logic.
  • [SAFE] (SAFE): No patterns of data exfiltration, hardcoded credentials, obfuscation, or unauthorized command execution were detected. The skill uses standard developer tools for its intended purpose of repository maintenance.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:30 PM