proof-gates

Reference implementation and pattern library for proof gates: agent-authored verification contracts resistant to sycophancy and distribution skew. Reproducibility artifact for the paper 'Proof Gates' (Whittaker 2026).

PVDM: proof-gated publication. Verify-before-commit content integrity for serverless data-mesh lakehouses (reference gate + validation suite).

33 specialized RIG agent skills for governed AI systems — deviation engines, doctrine, studios, engineering, fleet, knowledge, platform. Install in Claude Code, Codex, Hermes.

Operating doctrine agents load before they act — proof standards, gates, TAC rules. AI agent governance and deployment safety.

Make any AI repo 1000x governed — TAC doctrine, L10 self-evolving tests, L8 verification, OpenSpec BDD, ProofPates. One command: apply-overlay.sh

Identity-aware pre-commit linter. Validates references to specific people against a JSON-defined SSOT. Sibling artifact to proof-gate-patterns; demonstrates the operator-authored verification pattern from the paper 'Proof Gates' (Whittaker 2026).

Personal site for Mike Rodgers — legacy/predecessor version, being consolidated into mikerodgers-tech-v10.

Operator workstation shell — evaluated/adapted OpenWork (upstream: different-ai/openwork-workspace, MIT/Fair-Source) for RIG workflow review.

Prompt intelligence engine — deterministic 4-axis scoring (Specificity, Doctrine, Context, Actionability). Score, enhance, and fix prompts before they ship.

Boot configurator experience site — early asset stage, no implementation yet.

40 AI deviation engines × 14 sigma rungs — push LLM output past the generic median. Cognitive, nature, and physics layers for AI content generation.

npx mrodgersjs-web — install Mike Rodgers as a local Forward Deployed Engineer CLI teammate. Ask, audit, deploy. Zero dependencies.

Mike Rodgers — Forward Deployed Engineer resume in markdown, PDF, and docx. Enterprise AI solutions architect, deployment strategist.

GitHub-native agent firm — 6 role-agents (Verifier, Operator, Fleet, Builder, Strategist, Governor) with heartbeat telemetry and forkable constitution.

Gated communications protocol engine with scored formulas — AI-driven messaging with approval gates and compliance checking.

Agent role contracts (markdown) with Builder-Verifier separation — role definitions and boundaries as documentation, not runtime enforcement.

Operator OS with L10 self-evolving harness — 38 core tests, closed-loop verification, agent governance. Local-first AI operations platform.

Catch AI agents when they lie about done — HMAC-signed ProofPackets make completion claims cryptographically re-verifiable. Forward Deployed Engineer tool for governed AI production.

L10 self-evolving LLM evaluation harness that learns from failures — faithfulness, relevance, toxicity, format metrics with adaptive test generation.

LLM output validation with proof-gated completion — schema, toxicity, factuality, format checks with HMAC-signed ProofPackets. AI guardrails for production.