profiling-threat-actor-groups
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves the MITRE ATT&CK STIX database from the official GitHub repository (mitre/cti). This is a standard procedure for security tools and is considered safe.
- [COMMAND_EXECUTION]: The provided Python script (scripts/agent.py) is used for processing threat intelligence data and does not contain any arbitrary command execution or dangerous shell operations.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data. 1. Ingestion point: scripts/agent.py downloads the ATT&CK JSON bundle. 2. Boundary markers: The source is a structured data bundle; no explicit prompt boundary markers are used. 3. Capability inventory: The agent performs file operations in /tmp and specific network requests to GitHub. 4. Sanitization: The script uses the stix2 library which provides validation for STIX objects, and it only extracts specific metadata fields for display.
Audit Metadata