glean-incident-runbook

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill follows security best practices for diagnostic runbooks and incident response.
  • [DATA_EXPOSURE]: Authentication is handled correctly via the $GLEAN_API_TOKEN environment variable, preventing hardcoded credentials in the instructions.
  • [INDIRECT_PROMPT_INJECTION]: The skill interacts with external API data, creating a potential ingestion surface for indirect instructions. This is handled safely by filtering specific fields.
  • Ingestion points: JSON responses from Glean API endpoints (SKILL.md).
  • Boundary markers: Absent.
  • Capability inventory: Shell execution via Bash, network requests via curl (SKILL.md).
  • Sanitization: Present; uses jq to extract specific status and statistics fields from the JSON payload rather than interpreting the raw response as text.
  • [EXTERNAL_DOWNLOADS]: The skill communicates with Glean's official API domains for diagnostics, which is consistent with its stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 06:25 PM
Security Audit — agent-trust-hub — glean-incident-runbook