Most CRMs are a database with a form in front of it. The AI ones bolt a chat box onto the side of that form. Both leave the actual work — finding out what is true, and writing it down — to a human who has better things to do. This is built the other way round.
MIT · v1.2.0 · macOS 105 MB (Apple silicon) · Windows 129 MB (x64) · Chinese and English · carries its own database
The agent runs on its own deployment, on its own schedule, against its own work queue. It decides what to look at next, books its own follow-ups, spends a research budget, and stops when the budget runs out. Close the browser and it keeps going. Nothing about it is request-response.
No tool accepts a confidence score, because a model asked to grade its own certainty will, and it will be wrong in the direction that makes it look useful. Tools report what they observed — a signature block, a public developer profile — and a ledger prices the evidence. Strong evidence writes to the record. Weak evidence becomes a suggestion a human settles. A confidently wrong fact about a customer is worse than a blank field, because nobody can tell it is wrong.
It works with no API keys at all: it reads your own threads, meetings and signature blocks, which is free and is the best evidence there is — no data vendor can sell you a reply from the person’s own address. Each key opens one more place to look, and it is told at the start of every session which ones this install has, so it plans around what it actually has rather than discovering the gaps one failed call at a time.
The agent gets bash, grep and a workspace, with deny-all egress. The other half of that rule is an absence: the sandbox is never given a database URL. A shell with credentials and egress is exfiltration-shaped even in an internal tool; a shell with neither is a text processor.
Every contact, company and deal has an Agent tab — the steps as it takes them, the leads it throws away and why, and its questions answered in place when it cannot decide between two people. Conversations are durable and survive a reload.
Telemetry is off, and it has no default destination. The upstream analytics key that shipped with this codebase has been removed — software you run on your own hardware should not phone anywhere until you say so. Point it at your own PostHog project if you want the numbers.
The installer carries its own Postgres and its own JavaScript runtime. There is no database to provision, no Docker, and no account to create anywhere — open it and it builds the database itself.
macOS
1. Open the .dmg and drag OPENVZ CRM into Applications
2. The first launch needs a right-click → Open → Open
3. Give it about a minute to build the database. Only the first time.
4. Create the first account with an email and a password
Windows
1. Run the .exe. SmartScreen warns once: More info → Run anyway
2. The rest is the sameThe first account you create owns this CRM, and no second account can be created — to let somebody else in, open settings.env in the data folder, put their address in ALLOWED_SIGN_IN and restart (Google credentials go in the same file, if you would rather sign in that way). Neither installer is signed, so each system warns once. Everything stays on your machine: ~/Library/Application Support/OPENVZ CRM on macOS, %APPDATA%\OPENVZ CRM on Windows. Deleting that folder is a factory reset and it takes the data with it.
The CRM works after a relationship exists: getting what already happened written down, and moving it along. If you do not have a list yet — if the job is finding the people, working out whether they are worth talking to, and writing the first email — that is OpenVZ Leads.
See LeadsOPENVZ CRM is a derivative of an MIT-licensed codebase. The work went past renaming: the brand, icon set and colour system were rebuilt, telemetry was inverted from on-by-default to off with the upstream analytics key removed, and the upstream project’s customer logos and staff photographs were taken off the landing page. The original copyright notice stays in LICENSE, as MIT requires, and the list of changes ships with the source.
The installed copy brings its own database and works out of the box. The source is complete too; hosting it yourself needs your own Postgres and an OAuth client. If you want it reshaped around how your team actually sells, or simply hosted for you, say so.
Get in touch