Obizworks
Management PortalEnterprise Resource Management
Enterprise Resource Management

One dashboard to run the whole business.

Watch it, protect it, and build on it — from one screen. Logs, monitoring, backups, and a growing library of ready-made skills, tools and AI agents you can put to work.

Request a demoSee the dashboard
KNOW WHAT YOU HAVE KEEP IT SAFE AI AssistantEvery project at a glance— scope, status, paths CDC · WikiAsk your businessanything, get a source BackupsDaily, with one-clickrestore you can prove Password VaultCredentials locked away,access recorded SEE WHAT'S HAPPENING PROVE IT LATER Activity LogsWho did what, acrossevery portal you run MonitoringUptime, expiry andhealth — before it bites AI GovernanceEvery decision writtendown and dated Google WorkspaceDrive, Gmail andCalendar in one place THE LIBRARY — BUILD ONCE, REUSE EVERYWHERE SkillsAbilities your agentscan be given ToolsOne-click jobs, runsafely and logged Modules & Plug-insSwitch on a featurefor any portal AgentsDigital workers, eachon a signed contract
Live now Rolling out On the roadmap

Twelve modules, one login. You see everything your business is doing in one place — and nothing else needs a separate tool.

Four things it does for you.

You don't need to learn a system. Each part answers a question you already ask.

Know

"What have we got?"

Every project, its status and where it lives — plus a knowledge base you can simply ask questions of.

Protect

"What if something breaks?"

Daily backups of everything, restore in one click, and credentials kept somewhere safer than a spreadsheet.

Watch

"Is anything wrong right now?"

Uptime, expiry dates and a single activity log showing who did what, everywhere, without hunting.

Build

"Can we do that again, faster?"

A library of skills, tools, modules and agents. Solve it once; every future project starts from there.

You solve it oncefor one project It goes in the libraryas a reusable piece — named, versioned, safe Next project The one after that Your whole team Nothing is built twice. Every project starts further along than the last one.

This is the part that compounds — the reason year two costs less than year one.

The Obizworks hallmark — every portal in the estate runs on the same governed platform.

Safe agent development & hosting

How your agent is kept safe

When Obizworks develops or hosts an AI agent for you, it doesn't run loose. It runs inside a layered environment — every layer a control the system enforces, not a policy you have to trust.

Layered containment environment for a client agent Concentric layers around the client's agent: Constitution, identity and capability contract, budget cap and egress allowlist, append-only audit trail, with human override able to reach in at any layer. 1 · CONSTITUTION Ratified rules every agent loads on every run. Bedrock rules cannot be overridden — by any contract or person. 2 · IDENTITY & CAPABILITY CONTRACT A named agent that may do only what its signed contract declares — nothing more. 3 · BUDGET CAP & EGRESS ALLOWLIST A hard spend ceiling, and a list of the only endpoints it is allowed to reach. 4 · APPEND-ONLY AUDIT TRAIL Every action recorded — never edited, never deleted. The answer to "what happened?" YOUR AGENT does the work — inside all of it HUMAN OVERRIDE a person can pause or stop it at any time
1Constitution. The rulebook your agent is born into. Bedrock rules can't be weakened — not by a client contract, not by us.
2Identity & capability contract. A known, named agent, allowed to do exactly what its contract declares.
3Budget cap & egress allowlist. A hard ceiling on spend, and a fixed list of where it's allowed to connect.
4Append-only audit trail. Everything it does, written down permanently — so any action can be explained, later, to anyone.

Every layer is enforced by the substrate itself. Exceed a budget, reach an endpoint that wasn't granted, act outside the contract — the system stops it, and the attempt is in the audit trail. That's governance that holds, not governance you point to.

Technical infrastructure

Governance is only as solid as what it runs on.

The whole platform runs on a hardened Microsoft Azure virtual machine in West Europe — chosen for data residency, built-in security tooling, and a clean vertical scaling path. Every component below is a named, monitored part of the stack.

Obizworks technology infrastructure stack Public traffic enters through Cloudflare, then a hardened Microsoft Azure D2s_v3 virtual machine in West Europe running Caddy, the governance API, the metering proxy, the governance MCP, PgBouncer and PostgreSQL 16 with pgvector. Backups are GPG-encrypted to Azure Blob Storage and logs stream to Azure Monitor. FROM THE PUBLIC INTERNET CLOUDFLARE WAF · CDN · DDoS shield · TLS at the edge MICROSOFT AZURE VM Standard D2s_v3 · 2 vCPU / 8 GiB RAM · West Europe · Trusted Launch (Secure Boot + vTPM) obz-caddy reverse proxy · Let's Encrypt TLS · /v1/* + /governance/* routing governance-api Constitution · contracts · records :8000 · FastAPI obz-llm-proxy per-agent metering · budget caps · egress :8002 · Anthropic-compatible governance-mcp agent tool surface MCP server PgBouncer connection pooler — apps never hit the database directly PostgreSQL 16 + pgvector obz_audit · obz_workflow · obz_context — append-only audit substrate Azure Monitor Log Analytics syslog + alert rules obz-backup daily 17:00 UTC GPG-encrypted Azure Blob Storage off-VM · GPG-encrypted · tested restores
Standard D2s_v3 · 2 vCPU / 8 GiB West Europe region Trusted Launch · Secure Boot + vTPM Premium SSD · daily encrypted backups

Cloudflare absorbs hostile traffic at the edge; Caddy terminates TLS and routes only known paths; the governed app containers never touch the database directly — they go through PgBouncer into PostgreSQL 16 with pgvector. Backups leave the VM nightly, GPG-encrypted, to Azure Blob Storage, and the whole machine streams its logs to Azure Monitor. Scaling up is a VM-size change, not a re-architecture.

Common questions

What is the Management Portal, in one sentence?
It's the dashboard for your whole business — one login that shows what you have, keeps it safe, tells you when something's wrong, and gives you a library of ready-made pieces to build with.
Do I have to use all of it?
No. Each module works on its own and can be switched on when you need it. Most businesses start with backups and the activity log, then add the rest as they grow into it.
What's the "library" for?
It's where solved problems live. When something is built once — a way to deploy a site, a report, an AI agent that handles a task — it's saved as a reusable piece. The next project uses it instead of starting over. That's the difference between a business that gets faster and one that keeps paying for the same work.
What are "agents", and should I be worried about them?
An agent is a digital worker that handles a task for you. Every one runs inside a signed contract, a spending limit, and a permanent record of everything it did — and a person can stop it at any moment. See "How your agent is kept safe" above.
Is this only for large companies?
The opposite. Big companies have staff for each of these jobs. A small business has one person doing all of them — which is exactly why having them in one place matters more.
Can I prove what happened, months later?
Yes. Every action is written to a record that can't be edited or deleted afterwards. When a client, an auditor or an insurer asks what happened and when, you show them rather than remember.

More on how the whole governed platform works — obizworks.com/answers.

Obizworks Enterprise

Book a governance teardown of your AI stack.

Book a call ›
obizworks.com

Part of Obizworks Enterprise

Obizworks builds governed AI that runs your workflows — audited, human-approved, and accountable.