Lovrabet CLI 2.0
Lovrabet CLI is best understood as the business execution layer behind an Agent. Business users should not need to memorize commands. In most cases, the Agent, its skills, and the delivery team handle the CLI calls in the background.
April 21, 2026 Update
On April 21, 2026, Lovrabet CLI officially launched as an AI operation suite for business scenarios.
- Product page: https://www.lovrabet.com/products/lovrabet-cli
- Documentation: Lovrabet CLI 2.0
Lovrabet CLI is not a command-line copy of an admin console. It is the runtime operation entry that brings Lovrabet capabilities into the AI ecosystem. It turns app directories, datasets, data operations, SQL, BFF, and diagnostics into stable commands, so Agents, delivery teams, and business operations teams can work from real business objects such as customers, orders, inventory, and tickets.
Its business value is to turn enterprise data, rules, interfaces, and accumulated operating experience into business capabilities that AI can understand, call, audit, and reuse. A business request can then become a repeatable, controlled, and collaborative Agent workflow.
How this is meant to work
In a typical Lovrabet setup, there are two layers:
- Business users describe the business problem in plain language.
- Agents and skill maintainers decide which app, dataset, SQL, BFF, or collaboration CLI should be used behind the scenes.
So the goal is not “teach everyone the CLI.”
The goal is “give the Agent a stable, controlled, auditable way to execute business workflows.”
What it is good at
- Finding lists of customers, orders, tickets, projects, or inventory items
- Looking up a single record or tracing one business issue
- Producing grouped counts and summary numbers
- Reusing existing SQL and BFF logic
- Performing controlled write operations such as create, update, or delete
Why it is designed around Agents and skills
Lovrabet CLI is not trying to expose every internal capability directly to business users.
That is why you will see:
sql execandbff exec- but no business-facing
sql listorbff list
This is intentional. Sensitive business logic should normally be routed through a known skill, not explored freely by a generic agent.
The business user should say:
“What result do I need?”
The skill should answer:
“Which app, dataset, SQL, or BFF is allowed for this workflow?”
A typical flow
- A delivery or integration teammate installs the CLI, logs in, confirms visible apps, sets a default candidate if useful, and installs skills.
- A business user describes a real business request.
- The Agent uses skills to decide which Lovrabet capabilities are needed.
- The Agent runs
lovrabetand, when needed, also calls Lark CLI, DingTalk CLI, or other ecosystem CLIs. - The final output is a list, summary, recommendation, notification result, or a controlled write result.
How business users should ask
The most useful pattern is:
Help me in 【which business system or business domain】, find or process 【which business object】, within 【time range / warehouse / store / project / customer group】, under 【status or field conditions】, and give me 【a list / summary / recommendation / notification result】. This request is 【read-only / allowed to write】.
Examples:
- Help me find customers created in the last 30 days who have not placed their first order yet, group them by industry, and give me a sales follow-up list. Read-only.
- Help me find paid but unshipped orders from the last 7 days, group them by warehouse, and give me the exception order list. Read-only.
- Help me change ticket 241 to “processing.” Preview first, then wait for my confirmation before executing.
If you want more examples, read:
How it differs from Rabetbase CLI
You can think of the two CLIs as working in sequence:
- Rabetbase CLI is mainly for the development and delivery stage.
- Lovrabet CLI is mainly for the runtime stage: querying, validating, executing existing logic, and helping Agents run real business workflows.
In one sentence:
rabetbasehelps build capabilitieslovrabethelps run those capabilities in real business work
How to read the docs
If you are a business user or Agent consumer, start here:
- How to ask business questions so the Agent can actually help
- Common business scenarios
- Case study: a business user makes a request and the Agent delivers across CLIs
If you are an integration engineer, delivery engineer, or skill maintainer, continue with:
- Installation and login
- Dataset discovery
- Data operations
- SQL and BFF
- Advanced usage: config, cache, and integration troubleshooting
- Safety and automation
- Advanced tips and troubleshooting
- Lovrabet CLI risk levels
One final thing to remember
The real value of Lovrabet CLI is not that more people learn more commands.
The real value is that a business request can be turned into a stable, repeatable, and safe Agent workflow.