kovca CLI · macOS, Linux, and Windows

Build automations without leaving your terminal.

Describe a workflow to the Kovca agent, or edit it in a full-screen terminal Studio. Test with live step progress, publish immutable versions, and script it all in CI.

curl -fsSL https://kovca.io/install.sh | sh
Start freeRead the CLI docs
Agent mode

Describe it. The agent builds it.

Run kovca agent and say what should happen in plain words. The agent picks the nodes, wires them up, validates, and tests, showing every step as it goes.

  • Builds and changes workflows through the same validated edits as the canvas, on a draft you can open in the Studio.
  • Tests drafts on your examples with real Decision AI and Generative AI, while HTTP and app actions are simulated.
  • Asks before it publishes. In print mode (-p) it stops with exit code 7 until you confirm or pass --yes.
  • shift+tab opens the workflow it's working on in the terminal Studio, and back.
kovca agent
kovca agent "Build a workflow that screens return requests against our policy,
  asks #support-leads before anything over $500, and test it on three examples."

# In scripts: print mode, JSON out, and an explicit yes to publish
kovca agent -p --json --yes --workflow "Return screening" \
  "Also post in #returns when a label is created"
Terminal Studio

A visual Studio, in your terminal

Run kovca studio to see a workflow as a live graph with each node's settings beside it. Test it and watch the steps light up as they run, then publish without opening a browser.

arrows
walk the graph
enter
edit a setting
a
add a node
c
connect
T
test the draft
R
run the published version
p
publish
A
automation
/
ask the agent
shift+tab
Agent mode
Scripting and CI

Every action is a command

Set KOVCA_API_KEY and the same commands run in CI: data on stdout, progress on stderr, and exit codes you can branch on.

  • Same rules as the web

    The CLI edits the same drafts, runs the same validation, and publishes the same immutable versions as the Studio and the API.

  • Test before anything ships

    kovca test runs the draft like the Studio's Test: AI steps are real, and actions are simulated unless you pass --live-actions.

  • Workflows as code

    kovca workflows export and apply keep a workflow in your repository, so changes go through pull requests and CI.

CI
export KOVCA_API_KEY=kvc_…            # a secret in your CI system
kovca workflows apply workflows/return-screening.json --publish
kovca test "Return screening" --input tests/unworn-shoes.json --json > result.json
kovca run "Return screening" --input-json '{"order_id":"4821"}' | jq -r .decision

Prefer a canvas?

The web Studio edits the same workflows, with an assistant that builds them from a description.

See the Studio Use your own AI agent