White papers
Four papers on governing AI access to the systems your business runs on. Each is written for a different reader, each states what is built and what is only designed, and each ends with the limits rather than a summary. Start with the security paper; it fixes the vocabulary the other three use.
What is not in these papers
These papers summarize a larger internal documentation set: the full threat model with all ten scenarios written out, the audit events contract with its schemas, the tool-server contract every adapter is built against, the policy schema, and the decision log behind every architectural choice. Customers, prospective customers, and their auditors can request the whole set.
If you are evaluating us and want the adversarial review before the pitch, ask for the threat model. We wrote it before anyone asked.
Write to hello@omnafy.com.