glance-config

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill documents and references absolute filesystem paths belonging to a specific user environment. Evidence: Multiple references to /Users/wangyu19/Documents/Vibe-Coding/Glance-Monitor/ are found in SKILL.md and references/project-widgets.md. This reveals the local username and internal directory structure.
  • [INDIRECT_PROMPT_INJECTION]: The skill manages the configuration of widgets that ingest and render data from external API endpoints. Ingestion points: The custom-api and extension widget types in SKILL.md and references/glance-core-reference.md. Boundary markers: The skill instructions include directives to 'call out the trust risk' in SKILL.md. Capability inventory: The skill generates configuration files that enable the Glance dashboard to execute Go templates and render HTML, including a property allow-potentially-dangerous-html. Sanitization: The skill notes that allow-potentially-dangerous-html is disabled by default.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 05:29 PM