glean-multi-env-setup

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides legitimate configuration and workflow guidance for Glean environment management. It correctly utilizes environment variables (e.g., GLEAN_API_TOKEN_PROD) and .env files for secret management rather than hardcoding credentials.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes a promotion workflow that ingests local JSON files (test-docs.json, baseline.json, and connector configurations) which are then processed by the agent using curl and local scripts.
  • Ingestion points: The skill reads local data files in SKILL.md.
  • Boundary markers: None provided for file contents.
  • Capability inventory: The skill utilizes shell commands (curl) and script execution (node).
  • Sanitization: Not explicitly defined in the promotion snippets.
  • [COMMAND_EXECUTION]: The skill includes instructions for running a local Node.js script (scripts/compare-search-quality.js) and curl commands to interact with the official Glean API (*.glean.com). These are standard operations for the stated purpose of enterprise search management.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 06:26 PM
Security Audit — agent-trust-hub — glean-multi-env-setup