security-investigator-automation
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions require cloning a repository from
github.com/SCStelz/security-investigator. This external source is not a pre-verified vendor. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection due to its core function of processing untrusted security data.
- Ingestion points: The agent ingests data from Microsoft Sentinel logs, Defender XDR alerts, Entra ID activity logs, and various external threat intelligence APIs (IPInfo, AbuseIPDB, Shodan, VPNapi).
- Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the processed log entries.
- Capability inventory: The skill has access to execute complex KQL queries via MCP servers, run local Python scripts for report generation, and perform network requests for threat intelligence enrichment.
- Sanitization: There is no evidence of sanitization or filtering of the ingested security event data (such as process command lines or user-generated descriptions) before it is analyzed by the agent.
Audit Metadata