An AI agent inside your business,
on a stack you own
Connected to your CRM, your inbox and your files, it knows every customer, every job and every conversation, and works across all of them at once. Every call on the record, every quiet account noticed, every pattern nobody had the hours to find. On your accounts, with your choice of model.
How a project starts →
What an agent
is made of
Connected to your CRM, your inbox and your files, it knows every customer, every job and every conversation, and works across all of them at once. Three things underneath make that possible: a harness, connectors and skills.
Notices what nobody had the hours to find
Every Monday the review agent reads the CRM, the inbox and the call transcripts for every active account, and leaves a task on each one that has gone quiet. A minute of a person's attention per finding, instead of a week of reading.
The harness
Plans a task, works through it in steps, calls tools, holds context across a long job, checks its own output and recovers when something fails.
Connectors
Your CRM, Gmail, Drive and Calendar through their connectors, your own systems through the open connector protocol. Adding one is configuration, not a project.
Skills
Your procedures written down in plain text. How a quote is checked, what counts as quiet, what counts as finished. Change a line and the agent works differently on Monday. Someone non-technical can open and edit them.
Nothing
stops quietly
The jobs that run on their own, morning keys sync, overdue reminders, the settlement folders, each record their last run and its result. A job that stops is reported as stopped, not left showing as scheduled with nobody the wiser.
The failure that costs money is never the loud one. It is the job that has not run for three weeks while the dashboard said it had.
A bill you can
break down
Every call is recorded against the agent and the model that made it. You see which one is expensive, which one runs the most, and how much of a long task came from cache instead of being paid for again.
The live figures sit in your own provider accounts. A number nobody can break down is a number nobody can control.
Any model,
per agent
Claude, GPT, Gemini, DeepSeek and Kimi sit behind one routing layer on accounts in your name. Each agent gets the model that fits its job, a strong one for drafting, a cheap one for checking, with a fallback order for when a provider is having a bad day.
The model you would pick today is not the one you would pick in six months. Changing it is a settings edit, not a rebuild.
How a project
starts
Three steps before anything is built: a conversation about the jobs, a written scope, and one fixed price for the build. There is no list price, because no two agents do the same work.
Model usage is billed to your own provider accounts at the provider's list price, never marked up.
The conversation
What the business already has: the systems, the records, the calls and emails that nobody has time to read. What an agent takes over first, and who looks after it once it is running.
The scope
Written down before anything is built: the agents, the connectors, the schedules, what is in and what is out, and which accounts it all sits on.
The price
One figure for the build, fixed once the scope is agreed. What runs afterwards, the model usage and any ongoing arrangement, is listed separately so the two never blur.
Get in touch
Put an agent to work
Every business has work that nobody gets to: every call and email with a customer transcribed and on the record, the connections between accounts that no one person can hold in their head, the report that would take a week by hand. The first conversation is about which of those comes first. From there you get a written scope and a fixed price.
See the Content Engine→Questions people ask before they get in touch
An agent plans a task, works through it in steps, calls tools, checks its own output and recovers when something fails. A chatbot answers and stops. An automation follows fixed rules and gives the same result every time. An agent also does work no person has the hours for: reading every call and email with a customer and keeping it on the record, holding the whole history of an account in view, noticing what connects one customer to another.
It is quoted as a fixed price for the build, after a scoping conversation. The figure moves with the number of agents, the systems they have to reach and who maintains them afterwards. Model usage is billed to your own provider accounts at the provider's list price, never marked up.
Whichever fits the job, set per agent in configuration rather than in code. Claude, GPT, Gemini, DeepSeek and Kimi sit behind one routing layer on accounts in your name, with a fallback order for when a provider is unavailable. Swapping models is a settings change, not a rebuild.
Yes. Gmail, Google Drive and Calendar through their standard connectors, and your own systems through the open connector protocol. Any system with an API can be added as configuration rather than as a fresh build.
Every call is recorded against the agent and the model that made it, and reported monthly. The live figures sit in your own provider accounts, so you can check them any time without asking us.
No, but it changes the shape of the engagement. Where there is someone internal, they are usually the right person to build the workflows, because they know how the business actually runs. Our job is then the foundation plus a standing weekly session. Where there is nobody, we build the agents as well and hand over documentation and a walkthrough.
Every run is inspectable: what it was asked, which tools it called, what came back. Skills tell it how a job is done, so it is not improvising. Where it is unsure it flags rather than guesses, and anything consequential gets a human review point before it goes out.
Every scheduled job records its last run and its result. A job that stops is reported as stopped, not left showing as scheduled. The failure that hurts is the one nobody notices for three weeks, and the build is designed so that cannot happen quietly.
It runs on your infrastructure and your provider accounts, under commercial API terms that exclude training on your data. Where that is still not enough, private on-premise models are part of our AI practice. Credentials live in a password manager, never in code.
The code, the databases and the provider accounts are yours. It is built on standard tooling and open protocols, so another developer can pick it up. Nothing switches off and there is no licence to renew.