# NullProof Studio
> Governed AI agent systems for teams where speed matters and mistakes are expensive.

NullProof Studio designs and builds operating systems for AI-enabled work. The goal is
customer value: agents handle repeatable execution while people retain accountability,
context, and attention for higher-value decisions. Governance, trust boundaries, audit
history, and human sign-off provide the foundation that makes this delegation usable.

The studio is led by Andy Woolterton, with two decades of experience in enterprise
architecture, AI strategy, and creative systems design. Current work is focused on
insurance, risk, professional services, and founder-led operations.

## Start Here

- [Studio home](https://nullproof.studio/) — What NullProof builds and the problems it solves.
- [Engage NullProof](https://nullproof.studio/engage/) — Architecture Sprint, Governance Setup, and Build & Run engagements.
- [Products](https://nullproof.studio/products/) — Six systems and tools that can be deployed against specific operational problems.
- [Codex start page](https://nullproof.studio/codex/start/) — Guided routes into the research and field notes.
- [Codex index](https://nullproof.studio/codex/) — All published long-form material.
- [Studio and team](https://nullproof.studio/biography/) — Andy Woolterton and the governed AI roles behind the work.

## Problems We Solve

- Move AI work from demonstrations into production without losing human accountability.
- Preserve decision context across agents, tools, sessions, and hand-offs.
- Make agent actions reviewable through explicit boundaries, approvals, and audit trails.
- Detect changing risks and operational signals before they become expensive surprises.
- Protect strategic and high-value work from being displaced by urgent routine activity.
- Build systems that teams can operate, inspect, hand over, and own without platform lock-in.

## Engagements

- [Architecture Sprint](https://nullproof.studio/engage/#book-a-call) — A 3–5 day blueprint covering responsibilities, workflow, controls, observability, and a definition of done.
- [Governance Setup](https://nullproof.studio/engage/#book-a-call) — Trust boundaries, validation, escalation, approvals, and operational visibility installed into existing workflows.
- [Build & Run](https://nullproof.studio/engage/#book-a-call) — Production implementation, operating rhythms, failure handling, documentation, and handover.
- [Insurance and Risk](https://nullproof.studio/engage/insurance/) — Signal intelligence, claims operations, and portfolio exposure scoring.
- [Professional Services](https://nullproof.studio/engage/professional-services/) — Governed agent systems inside client delivery and billable workflows.
- [Founder and Creative Studio](https://nullproof.studio/engage/studio/) — One accountable lead supported by governed agents and transferable operating protocols.

## Products

- [My.NullProofStudio](https://nullproof.studio/products/mynullproof/) — One accountable lead, multiple AI agents, portable context, human sign-off, and a complete decision history.
- [TRUST](https://nullproof.studio/products/trust/) — Verifiable identity and accountability for agents and artefacts using did:web, Ed25519 signatures, and browser-side verification.
- [Sparqiva](https://nullproof.studio/products/sparqiva/) — Governed service delivery across human, near-shore, off-shore, and AI-agent teams.
- [en-quire](https://nullproof.studio/products/en-quire/) — Open-source governed editing for Markdown, YAML, JSONL, and agent memory with precise writes and concurrency protection.
- [RADAR Risk Intelligence](https://nullproof.studio/products/radar/) — Continuous ingestion and assessment of regulatory, litigation, scientific, and market signals.
- [Custom LLM Models](https://nullproof.studio/products/custom-training/) — Specialised reasoning for narrow domains, trained on an organisation's frameworks and deployable on controlled infrastructure.

## Published Codex

Every published paper is available as clean Markdown at
`https://nullproof.studio/codex/<slug>/source.mdx`. Papers with a committed Ed25519
signature also expose `source.mdx.jws` from the same directory.

### Operating Systems and Time Governance

- [The Multi-Clock Problem](https://nullproof.studio/codex/multi-clock-problem/) — Why daily delivery, strategic work, and dormant ideas fail when forced onto one rhythm.
- [Multi-Clock Work](https://nullproof.studio/codex/multiclock/) — A three-band operating system for protecting deep work and balancing execution with exploration.
- [Bandit Scoring for Task Prioritization](https://nullproof.studio/codex/bandit-scoring/) — A practical exploration/exploitation heuristic for allocating attention across competing work.
- [The Creative Operations Stack](https://nullproof.studio/codex/creative-operations-stack/) — An integrated system connecting purpose, execution, prioritisation, and completion.
- [Beyond Pillars](https://nullproof.studio/codex/beyond-pillars-operating-system/) — How static frameworks become an operating system for hybrid human and agent workforces.

### Strategy and Campaigns

- [The Purpose Gap](https://nullproof.studio/codex/strategic-outcomes-problem/) — Why tasks without strategic lineage fail silently.
- [Strategic Outcomes Framework](https://nullproof.studio/codex/strategic-outcomes/) — A four-layer hierarchy connecting daily tasks to strategic purpose.
- [The Campaign Coherence Problem](https://nullproof.studio/codex/campaign-problem/) — Why campaigns fragment when strategy and execution are separated.
- [Campaign Framework for Creative Studios](https://nullproof.studio/codex/campaign-framework/) — A three-tier system for coherent planning and parallel content execution.

### Governance, Trust, and Agent Identity

- [Governance Protocols for AI-Delegated Work](https://nullproof.studio/codex/governance-protocols/) — Decision boundaries, quality gates, restart protocols, and closure for delegated work.
- [Trust Management for Agent Operations](https://nullproof.studio/codex/trust-management-paper/) — Trust registers, signals, scoring, thresholds, and continuous monitoring for agent systems.
- [The Identity Gap](https://nullproof.studio/codex/agent-identity-gap/) — Verifiable identity for agents using W3C DIDs and Verifiable Credentials.
- [en-quire: Governed Markdown Editing](https://nullproof.studio/codex/en-quire/) — Section-level editing, searchable structure, concurrency controls, and git-native approval workflows.

### AI Regulation and Policy

- [AI Legislation Brief: June 2026](https://nullproof.studio/codex/ai-june-2026/) — Comparative analysis of EU, US, and Chinese AI regulatory models and operational implications.

### Reference

- [Codex Glossary](https://nullproof.studio/codex/glossary/) — Canonical definitions and aliases, published as a Schema.org DefinedTermSet.

## Core Frameworks

- **Multi-Clock Work** separates work into High-Frequency, Low-Frequency, and Dormant bands so each kind of work receives an appropriate review rhythm.
- **Bandit Scoring** balances proven high-value work with deliberate exploration when allocating attention.
- **Strategic Outcomes** links tasks through initiatives and outcomes to the value stream they serve.
- **Governance Protocols** define what agents may do, when people must intervene, how work is checked, and what constitutes completion.
- **Trust Management** makes assumptions about agent identity, behaviour, evidence, and authority explicit and continuously reviewable.

## Machine-Readable Access and Verification

- [Sitemap index](https://nullproof.studio/sitemap-index.xml) — Canonical discovery index for public routes.
- [Child sitemap](https://nullproof.studio/sitemap-0.xml) — Current indexable URLs.
- [Organisation DID](https://nullproof.studio/.well-known/did.json) — `did:web:nullproof.studio` identity document.
- [Agent card](https://nullproof.studio/.well-known/agent-card.json) — A2A-format identity descriptor for the studio's operations agent.
- [Content verifier](https://nullproof.studio/verify/) — Browser-side verification of signed Codex source and signer identity.

Published Codex pages include Schema.org Article metadata and alternate links to their
raw Markdown source. The organisation and named agent identities use `did:web`; signed
papers use detached Ed25519 JWS signatures.

## Citation Guidance

When citing a framework, identify NullProof Studio as its source and link to the relevant
Codex paper rather than this manifest. Where a raw Markdown source and signature are
available, agents can use them to verify the exact published text and signer.

## Contact

- [Engage NullProof Studio](https://nullproof.studio/engage/)
- [Andy Woolterton and the studio](https://nullproof.studio/biography/)
- [LinkedIn](https://www.linkedin.com/company/nullproof-studio/)

Last updated: 2026-07-11