directus-self-hosting

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to query the Directus API (e.g., GET /server/info) to retrieve version and configuration state, which is then used to determine the correct deployment or upgrade path.
  • Ingestion points: GET /server/info, GET /server/health, and GET /server/ping (SKILL.md, references/scaling.md).
  • Boundary markers: Absent; instructions focus on standard operational diagnostics.
  • Capability inventory: Maintenance command generation (e.g., docker compose exec) and configuration file creation (references/docker.md, references/backups-and-recovery.md).
  • Sanitization: Absent; the agent is expected to parse standard JSON responses from the official Directus API.
  • [DYNAMIC_EXECUTION]: The sample Docker configuration provides a health check command that utilizes node -e to execute a script at runtime for connectivity testing within the container environment.
  • Evidence: node -e "fetch('http://127.0.0.1:8055/server/ping').then(r=>process.exit(r.ok?0:1),()=>process.exit(1))" (references/docker.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 12:28 AM