Most autonomous AI pentesters send your code, IPs and traffic to a hosted model. Darkmoon can run entirely on a local LLM, with a Privacy Gateway that keeps sensitive values away from the model, so it is safe to point at production or a client environment.
Local model, tokenized data, readable source, the wedge the high-star cloud tools do not have.
Run against Ollama or llama.cpp on your own hardware. The 57-finding OWASP Juice Shop benchmark was produced black-box on a local model, proof the local path works.
Learn moreWith a local model and the Privacy Gateway, no source, no traffic and no credentials are shipped to a third party. This is a design guarantee you can read in the code.
Web, API, Active Directory & identity, Kubernetes, cloud, CI/CD, databases, IoT/firmware and LLM endpoints, 50+ specialist agents, all running on your infrastructure.
The offensive benchmark and the 42/57 Pro remediation dossier publish their limits and invite you to reproduce or contest them, black-box, on a public lab.
Autonomous testing without a cloud model, and why it matters.
The paid Pro tier takes 42 of 57 findings to a fixed and retested end to end result, limits disclosed.
The black-box, local-LLM run, reproduce or contest every number.
Darkmoon is open source (GPL-3.0), self hosted, and runs on a local model. Clone it and read every line.