managing-test-environments

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface identified. The skill is designed to read and process configuration files from {baseDir}/config/, which could contain malicious instructions meant to influence agent behavior during test execution.\n
  • Ingestion points: {baseDir}/config/ (SKILL.md)\n
  • Boundary markers: Absent\n
  • Capability inventory: Bash(test:env-*) (SKILL.md)\n
  • Sanitization: Absent\n- [COMMAND_EXECUTION]: The skill utilizes a bash tool to invoke test frameworks. While providing significant capability, the risk is mitigated by an allowlist constraint requiring commands to be prefixed with test:env-*, preventing arbitrary command execution.\n- [EXTERNAL_DOWNLOADS]: The example test scripts suggest the installation of standard industry drivers (psycopg2-binary, redis). These are expected dependencies for a test environment manager and are sourced from official registries.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 11:42 AM