.lmi — Install | LMX

LMX // .LMI

.lmiInstall

Ask the RUNNING deployment what it is and what it can actually do.

  • Needs an install
  • a .lmi file
  • free to run
WHAT IT IS

Why .lmi exists.

Leumas is a product other people run, so "Leumas can do X" and "you can do X" are different claims. LMI asks the live instance instead of guessing: its tenant, role and edition; every kind it indexes (apps, tools, pages, modes, configs, capabilities and anything else wired in); whether THIS user is entitled to a capability and up to what limit; and which connected provider would actually fulfil a vendor-neutral outcome. Its answers are three-state — yes, no, and unknown — and it never reports unknown as yes.

Needs an install

Its vocabulary is a RUNNING Leumas install — your collections, your tools, your models. That is why it cannot run anonymously: with no account there is nobody for it to act as.


A WORKED EXAMPLE

What is this deployment, and what does it actually index?

The same script the engine test suite runs, so it cannot quietly stop working.

# What is this deployment, and what does it actually index?
INSTALL
KINDS

LIST apps LIMIT 5
FIND "chat"

1 more example

Reference: the report to run before telling a user what they can do here.
# Reference: the report to run before telling a user what they can do here.
INSTALL
KINDS

LIST tools LIMIT 10
LIST automations LIMIT 5

# Entitlement and fulfilment are DIFFERENT questions — a user can be entitled to
# a capability this install has no connected provider for.
CAN ai.text.generate
PROVIDERS FOR ai.text.generate

ORPHANS

SYNTAX

The shape of it.

Straight from the interpreter catalogue — the same summary the editor shows as a placeholder.

INSTALL
KINDS

LIST apps LIMIT 10
FIND "invoice"

CAN ai.text.generate
PROVIDERS FOR email.send

NEXT

Where to take it.

Or run it against an install you already have: the CLI takes --studio <url> --token <t>.