data-import-workflow

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill describes an ingestion point for external data from the tarkov.dev GraphQL API. 1. Ingestion points: Data retrieved via 'task importer:start' (SKILL.md). 2. Boundary markers: None explicitly mentioned in the documentation. 3. Capability inventory: Data is written to a database ('Persist') and local cache files. 4. Sanitization: The skill mentions a 'Transform' phase where external schemas are mapped to internal structs ('models'), providing structural validation. As this ingestion is the primary purpose of the skill, the risk is considered negligible.
  • [Data Exposure & Exfiltration] (SAFE): The skill references the use of a .env file for configuring 'POSTGRES_HOST'. This is a common and safe practice for local development, and no commands exist to exfiltrate this configuration to external servers.
  • [Command Execution] (SAFE): The use of the 'task' runner is appropriate for executing local build and import scripts mentioned in the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:48 PM