White Papers
Technical deep-dives on the cryptographic, operational, and governance primitives that make AI-driven software delivery defensible. Written for security leaders, compliance teams, and anyone evaluating AI-SDLC for regulated environments.
- RFC-0042v1.02026-05-21
Proof of Execution: Verifiable Evidence That Your AI Reviewers Actually Ran
Signed, tamper-evident evidence that a real review ran against the exact code, and why it matters for AI-driven delivery.
For: Security, compliance, and governance leaders evaluating AI-assisted development
Read the paper - RFC-0043v1.02026-06-03
Zero-Trust Untrusted-Contributor PR Verification
A four-stage zero-trust gate that lets automation review and merge pull requests from untrusted contributors — human or agent — without letting them escalate privilege, exfiltrate credentials, forge the attestation, or hijack the reviewers. Composes with RFC-0042 Proof of Execution.
For: Open-source maintainers, platform & security leaders
Read the paper