skills/mukul975/anthropic-cybersecurity-skills/detecting-lateral-movement-with-zeek/Gen Agent Trust Hub
detecting-lateral-movement-with-zeek
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs legitimate security analysis on local Zeek log files. All scripts use standard libraries and do not perform network operations or access sensitive system credentials.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Zeek network logs.\n
- Ingestion points: Reads
conn.log,smb_mapping.log,smb_files.log,dce_rpc.log,kerberos.log, andntlm.loginscripts/agent.pyandscripts/process.py.\n - Boundary markers: Standard delimiters (TSV/JSON) are used; no specific injection-prevention headers are present.\n
- Capability inventory: Limited to local file reading and console output. No network transmission or arbitrary command execution capabilities are exposed.\n
- Sanitization: Parsing is performed using structured formats (JSON/TSV), minimizing the risk of instruction leakage to the agent logic.
Audit Metadata