Original Research

Engineering and methods artifacts authored at RRM Academy and published as citable research. Each documents a verification discipline we run in production to keep AI-assisted work accurate.

Artifacts

  • proof-gate-patterns

    A pattern library for skill authors who want to add agent-authored proof gates to their LLM agents: deterministic verification code the agent writes per task, runs against its own work product, and quotes as proof, with fix-before-pass-through semantics at a workflow boundary. Reproducibility artifact for the paper Proof Gates (Whittaker 2026).

  • lint-identity

    An operator-authored pre-commit linter that validates references to specific people (canonical name, byline format, credentials, identifier aliases) against a single source of truth. The operator-authored sibling to agent-authored proof gates, on the same observable-output-as-proof axis.