B2B software companies
Customers ask how our agents are governed. Can we show them?
You build agents into your product and ship them to enterprises. Your customers now ask who owns each decision, why the agent acted, what it costs and where the record is. LeafCraft embedded in your product answers all of it, in their environment, under their rules.
What your customers ask
They did not ask for smarter AI. They asked for control.
Seven questions enterprise buyers now put to every software vendor. Each one is answered by LeafCraft embedded in your product, in their environment.
Who owns the handover?
When your agent passes work to their person, someone must own the result.Why did the agent act?
Logs are not enough. They want the reason, and what their person did next.Small problem or big, same person?
Escalation by complexity and impact to the right people. Not a ticket queue.Whose rules apply?
Theirs. Not your defaults.What does it cost, before it runs?
Not a bill after the fact.What happens when the model changes?
Switching should take a setting, not months.Can we export the history?
With links and reasons intact. Not a flat spreadsheet.The vendor builds the agent. The customer stays in charge.
How embedding worksThe agents you ship, and the agents you run.
Embed LeafCraft in your product
Your agents run inside each customer's LeafCraft, under that customer's rules, with the record in their database and the cost visible to them before it runs. You ship one product. Every customer governs it their way.
Embedded pricingYour own operations
Sales, support, finance and engineering at a software company run agents too, from vendors and from your own teams. The same layer governs them: one register, one set of rules, one record.
Discover every agentTwo risks the layer removes
The agents you can't see, and the model that gets retired.
Agents you don't know about
In B2B software companies, agents arrive through vendor updates and team pilots. LeafCraft finds all of them, gives each an owner, and brings it under your rules or removes it.
A model that gets retired
An agent built straight on one model has to be rebuilt and re-approved when that model is retired or repriced. On LeafCraft you switch the model for that agent in config and keep everything else. Months and a budget, saved.
- The model is retired or repriced
- Prompts, tests and sign-offs rebuilt
- Months of work, a new approval cycle
- The history stays with the old tool
- Pick another model for that agent
- Switching is a config change
- Test it against your own record first
- Rules, limits, owner and memory stay
Why it matters now
Every enterprise deal now has a governance question in it.
Your roadmap now has agents in it, and every enterprise deal now has a governance section in it. Each customer wants its own approval flow, its own limits and its own record, and each one you build by hand is engineering time that does not ship features. When a model is repriced or retired, the rebuild lands on your release plan, not theirs. LeafCraft embedded in your product turns governance into part of what you ship: customers set their own rules, hold their own record and see cost up front, while you build one product, switch models per agent, and close enterprise deals without a custom control plane per account.
The core infrastructure
What LeafCraft runs for the agents you ship.
One team
Your agents and your customer's people work as one team, in the customer's roles. Any seat can be your agent or their person, and no one checks their own work.
How it works02Your rules
The customer writes the rules: what your agent may do alone, what needs their sign-off, where it escalates. You ship one product that respects every customer's rules.
How it works03Your memory
Every decision your agent makes lands on a record in the customer's own database, with your agent's reason and their person's reason, exportable in open formats.
How it works04Your budget
Each customer sees the cost of your agents before work runs, with limits per task and per agent. When a model changes, you switch it per agent, not per release.
How it worksAI and people, one team
Your customer's team, with AI on the org chart.
The agent prepares the case. Your person makes the call and writes why. A separate audit agent checks the decision against your rules.
What changes for you
How it helps your people
Who it helps in B2B software companies.
Leaders who ask first
Functions where it runs
Questions teams ask
Can we use LeafCraft for the agents we ship, not just our own operations?
Yes. Agents built in LeafCraft or on other platforms run under the same rules, with a budget and a record. Each tenant's sessions, memory and events are stored apart from every other tenant's.
Do we have to rebuild our agents?
No. Bring them as they are. LeafCraft connects agents built on LangGraph, CrewAI, the OpenAI Agents SDK, the Claude Agent SDK and more, or over A2A, MCP or a web address.
Will our customers' text show up in traces?
Not if you turn on masking. One setting keeps message content out of traces, while timings, costs and agent names stay visible. One call erases a session on request, and the erase itself is recorded.
Bring one B2B software process.
We'll map it with you: which roles are AI and which are people, who decides, who checks, what it costs and what gets recorded.