Objects, fields, flows, validation rules and permission sets. Describe the change in plain language; Forge builds it in a sandbox and deploys it when you approve the change set.
Each area runs on the same engine. Most customers switch one on and add the rest as other teams ask for them.
Custom objects, fields, record types and page layouts, built from a description rather than a ticket.
Flow Builder logic with the validation rules and error paths that usually get written later, or never.
Profiles, permission sets and sharing rules, with the access change stated plainly before it is applied.
Everything built and exercised in a sandbox. Production sees a change set, not an experiment.
What changed, why, and what it touches, written as it is built, not reconstructed at audit time.
The small requests that queue behind the big ones, cleared continuously instead of quarterly.
The ontology is what separates a coworker from a chatbot. Forge holds your Salesforce build world as objects it can reason over and act on, not documents it searches.
Each one gated on a named human.
Derived live from the systems of record.
A sentence in, a reviewed change set out.
The existing object model, flows and permissions read into the ontology.
Dependencies mapped before anything is written.
What breaks, what is affected, what needs a test.
Working, in an org you can click through yourself.
Packaged and held. An admin signs it off.
A code assistant stops at step 3. Forge goes all the way to step 5.
Your org, your sandboxes, your release process. Forge works inside them rather than around them.
Each step of the loop is a separate worker from the Onetab Store, with its own tools and its own guardrail. That's why Forge can be audited step by step, and why a failure in one stage doesn't silently corrupt the next.
Forge belongs to a person, not a department. That person approves, and their name is on the record.
Every irreversible action pauses and waits. Forge proposes; a person decides.
Forge sees what the approver is allowed to see, and no more.
A per-agent budget, and every run stored and replayable for whoever asks.
Still deciding? Book a demo and we'll walk through Forge on your own tools.
No. It builds in a sandbox and prepares a change set. A named admin deploys it.
Impact analysis runs before the build, and the sandbox is where it is proven.
It clears the queue of small changes so the admin works on the ones that need judgement.
It reads the existing org and matches what is already there.
Sales & CRM
Follow-ups, CRM hygiene and routing qualified enquiries.
Works across
Software delivery
Triage, fixes and releases across your pipelines.
Works across
Forge is coming soon
Book a demo to see Forge early and help shape what it does first.
