automating-ioc-enrichment
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data such as IP addresses and file hashes from SIEM alerts and phishing emails to perform automated enrichment. This creates a surface for indirect prompt injection or path manipulation. Ingestion points: IOC extraction workflow in SKILL.md. Boundary markers: None present in the provided instructions or code. Capability inventory: Network request capabilities via the Python requests library and SOAR platform integrations. Sanitization: The Python code snippets do not include explicit input validation or escaping before interpolating variables into API URLs.
- [EXTERNAL_DOWNLOADS]: The skill makes network requests to well-known technology services and security providers including VirusTotal, AbuseIPDB, and Shodan. These are legitimate sources for threat intelligence data.
- [DATA_EXFILTRATION]: Indicators of compromise (IOCs) are transmitted to external enrichment services. This behavior is the primary intended function of the skill and does not involve sensitive local file access.
- [COMMAND_EXECUTION]: The documentation references platform-specific CLI commands (e.g.,
!vt-ip-scan) for the Cortex XSOAR platform. These are provided as instructional examples for building playbooks and are not executed in the local shell environment.
Audit Metadata