Every CLI capability the Connector exposes. Same binary that powers the local MCP server — every tool your AI assistant can call has a direct CLI counterpart.
$
kuali [global flags] <resource> <action> [flags] [arguments]
run any parent without a subcommand to see its help — e.g. kuali apps
$
kuali setup [--hostname <host>] [--api-key <key>] [--profile <name>]
$
kuali auth login | logout | status [--profile <name>]
$
kuali config set|get|list <key> [--profile <name>]
$
kuali apps list | get | create | publish | add-workflow-step
$
kuali documents list | get | create | update | submit | approve
$
kuali forms schema | options | outline | update
$
kuali users / groups / workflows / products / integrations …
$
kuali import csv · kuali export csv|pdf|xlsx
$
kuali mcp setup | verify · kuali doctor · kuali update install
The full reference lists every subcommand and flag per resource — the live page runs 20+ sections deep. This layout shows the pattern; extend with more resource sections as needed.
$
kuali documents list --app X -o json | jq '.[].id'
$
kuali documents list --app X --fields id,data.title --all
$
kuali documents delete <id> --dry-run
ON THIS PAGE
<- ReferenceThe server enforces your key's permissions. "Not found" or "forbidden" is usually a permissions issue, not a CLI bug.
First connection ->
Ready-to-use prompts for curriculum, research, build apps, imports, workflow analysis, and reporting.
Browse prompts →
Tips for Claude Desktop, Claude Code, Codex, Gemini, Copilot, and VS Code.
AI assistants →
Skip the chat — every capability is also a plain kuali command, ready for scripts and CI.
Command reference →
