Cisco-Talos/EvidenceForge ? reverse-engineered prompt

Build me a Python tool that generates realistic synthetic security logs from a YAML scenario file, so I can use it for threat hunting training and research.

I want it to take a scenario that describes users, hosts, network setup, and an attack story, then produce believable logs across multiple sources like Windows security events, Sysmon, Zeek, syslog, web, proxy, and IDS. The logs should make sense together, with events in a realistic order and timing, so things like DNS, connections, logons, and admin actions line up naturally. It should also support validating the scenario first, generating the output into a folder, and evaluating the result for realism and consistency.

Please include a simple command line interface, clear docs, and a few example scenarios I can run right away. If you need to check current docs or best practices online, go ahead. Make the output deterministic when given the same seed, and make it easy to customize personas, domains, applications, and other config data without breaking cross file references.