The six-layer framework for governing AI agents
A six-layer framework for governing AI agents: agent identity, delegated authority, policy at every tool call, accountability, audit evidence and security.
Last reviewed: 28 September 2026.
Comonad Limited · AI R&D and products
AI R&D and products for production systems
Comonad develops its own products and does applied R&D on AI in production systems: agent harnesses, the SDLC, automation and governance.
Our products
All productsProxed.AI
Secure AI API access for iOS apps, without building a backend.
- Split-key proxy for AI provider keys
- Requests verified by Apple DeviceCheck
- Hosted or self-hosted, AGPL-3.0
Plop
Stores and normalises inbound email and makes it available through a UI and an API.
- For operations, workflows and tests
- Open source under AGPL-3.0
- Source on GitHub: plop-email/plop
Notiterms
Monitors the published terms of the services you rely on, so the check runs continuously.
- Alerts compliance and commercial teams
- Feeds a continuous compliance process
- Operated by Comonad Limited
Applied R&D
Our R&D is for our own use. We work on what decides whether AI can run in a system that already has users, audit logs and a release process, and the results shape our products and our guides.
- Agent harnessesOrchestration, tool permissions, evaluation and guardrails around the model.
- AI across the SDLCA quality gate at each step, from specification to release.
- AutomationA schema for each output, a check that rejects what does not conform, and a route to a person.
- AI governance in productionIdentity, delegated authority, accountability, audit and security for AI agents.
Writing
All writingHow to govern AI agents in production
A practical guide to governing AI agents in production: tool permissions, evaluation, audit trails, human oversight, stopping safely and framework mapping.
Last reviewed: 28 September 2026.
Does the EU AI Act apply to UK companies?
When the EU AI Act reaches UK organisations, which duties already apply, the 2027 and 2028 high-risk dates, and how UK rules fit. Plain-English guide.
Last reviewed: 28 September 2026.
Alex Vakhitov
Founder, Comonad Limited · AI and software architect
- Cloud KB
- Co-founder & CTO, Cloud KB · 2015 to 2022
- SMETS2
- Supported SMETS2 installations from the early stages of the roll-out
- 2022
- Cloud KB acquired by private equity · 2022
Changing systems that cannot go down is slow, checked work: know the boundary, prove the new path, keep a way back.
Selective engagements
Work with us
Alongside our R&D and products, we take on a small number of client engagements, each agreed individually and led by Alex Vakhitov.
Who we take on
- AI in large production systems, operated like any other change
- UK enterprises and regulated organisations running AI
- Problems that match our R&D: agents, AI in delivery or governance
AI agents · ISO/IEC 42001
AI governance
Written controls for your AI agents and AI systems that your risk team can evidence.
Agent harnesses · Automation
AI engineering
Agents and automation built into the systems and release process you already run.
Questions about Comonad
What does Comonad Limited do?
Comonad Limited is a London company that develops its own products and does applied R&D on AI in large production systems. The R&D covers AI agent harnesses, AI in the software development lifecycle with quality gates, automation of engineering and business workflows, and governance for AI in production, mapped to the EU AI Act, UK regulatory guidance, ISO/IEC 42001 and the NIST AI RMF. We operate three products, Proxed.AI, Plop and Notiterms, and take on a small number of client engagements.
Which products does Comonad operate?
Comonad Limited develops and operates three products. Proxed.AI is a proxy that lets iOS apps call AI models without shipping full API keys or building a backend, verifying requests with Apple DeviceCheck; it keeps full AI provider keys off iOS devices. Plop stores and normalises inbound email and makes it available through a UI and an API. Notiterms monitors the published terms of the services an organisation relies on and notifies compliance and commercial teams when they change. Proxed.AI and Plop are open source under AGPL-3.0.
What does Comonad's R&D cover?
Comonad's R&D is applied and for its own use: agent harnesses, AI across the software development lifecycle with a quality gate at each step, automation with structured outputs and defined failure modes, and governance for AI in production, including a six-layer framework for governing AI agents. Alex Vakhitov leads it. The results shape Comonad's products and appear in its published guides.
Who is behind Comonad?
Alex Vakhitov, a London-based AI and software architect, founded Comonad Limited in May 2024. From August 2015 to March 2022 he was co-founder and CTO of Cloud KB, which built software for UK energy-market participants and supported SMETS2 installations from the early stages of the roll-out. Cloud KB was acquired by private equity in 2022. Alex leads Comonad's R&D and products, and the client engagements it takes on.
Where is Comonad based?
Comonad Limited is based in London, United Kingdom. It is a private limited company registered with Companies House under company number 15713725, with its registered office at 71-75 Shelton Street, Covent Garden, London WC2H 9JQ. It was incorporated on 10 May 2024 and is registered for VAT under number GB 519 2326 95.
Write to us about the products, the R&D or an engagement
A short note is enough. Choose a topic on the contact page.
71-75 Shelton Street, Covent Garden, London WC2H 9JQ