mapping-mitre-attack-techniques
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads the MITRE ATT&CK Enterprise matrix STIX bundle.\n
- Source: https://raw.githubusercontent.com/mitre/cti/master/enterprise-attack/enterprise-attack.json\n
- Purpose: To provide the base knowledge for mapping techniques.\n
- Assessment: The download targets a well-known service provided by MITRE, an established cybersecurity organization.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied detection rules and SIEM alerts, which serves as an ingestion point for external data.\n
- Ingestion point: The scripts/agent.py script reads detection rule data from the detection_rules.json file.\n
- Capability inventory: The skill is limited to generating reports and JSON visualization layers for the ATT&CK Navigator. It does not possess capabilities for code execution, network exfiltration of the processed data, or file system manipulation outside of its local directory.\n
- Boundary markers: The skill does not implement specific boundary markers or delimiters to isolate ingested rule logic from its own instructions.\n
- Sanitization: Data is parsed using standard JSON libraries, which mitigates simple control character injection, though the agent remains theoretically susceptible to logical prompt injection contained within the processed text fields.
Audit Metadata