Kalokoby

All guides › By use case

By use case

Design reference: build and test against the design

An accepted design version is the reference: agents build from its live files and tokens, and every run is checked for pixels, structure and tokens.

Once the team accepts a version of the design, it becomes the design reference of the scenario. Development builds from it and QA checks against it, step by step, with the same step ids.

The problem

Handoff files go stale the day the design changes. Nobody notices a button one shade off, a missing label or padding outside the scale until a designer reviews production by eye.

What changes with Kaloko

How it works

  1. Accept a version

    in the canvas, then *Set as design reference* (or kaloko reference --design --version 4).

  2. Build from it.

    Your agent calls get_step for each step, or runs kaloko export --live, and implements the steps with the project's own skills.

  3. Turn on the pack

    in the implementation scenario: packs: [design] on the steps that have a design.

  4. Run as usual.

    kaloko walk, kaloko evaluate, kaloko share. The evaluation fetches the reference and fills the design criteria.

  5. Follow changes.

    When the reference moves, list_step_changes names the steps to touch, and followers get *reference changed* in their inbox.

Skills and prompts

Implement the checkout steps from the design reference: get_step for each step of FLOW-checkout, keep the
step ids and the token names, then run the implementation scenario with packs: [design] and fix the drift.
List the steps that changed between the design reference and the draft, and plan the implementation work.

What it looks like

An implementation run shows AC801–AC803 per step and viewport. For example, *5 values outside the token set: padding 15px → space.m; background #2e6bfe → color.accent*, or *missing actions: "Pay"*. Compare offers the design reference next to the baseline.

What you get

More guides

Accepting a task with an AI agent, in the pull requestRegression before a release: compare the run with the accepted baselineSEO and landing-page checks on production, read-only, every dayE-mail flows: capture the message next to the screen that sent it

Install once, then work through your agent

Kaloko runs where your code and your agent are. The service stores and versions the results, shows the canvas and collects approvals.

  1. Add the CLI to the project
    npm install --save-dev kaloko

    Needs Node 20 or newer. Update later with npm update kaloko.

  2. Create the config and install the skill
    npx kaloko init --agent claude --org <your-org>

    The skill is copied to .claude/skills/kaloko. npx kaloko doctor checks Chrome, the config and the token.

  3. Create your organization and a token

    Create an organization; you become its admin. The start page offers a tester token in one click, later under Settings → API tokens. Put it into the project .env:

    KALOKO_TOKEN=qwk_…
    TYPESAFE_API_KEY=…   # optional: semantic evaluator

Then just ask your agent

The skill teaches your agent the whole loop: it writes the acceptance plan and the scenario from the task, walks the screens, evaluates, shares the canvas, reads what reviewers said and fixes it. You don’t type the commands; you look at the canvas.

What the agent runs (or run it yourself, e.g. in CI)

The same loop by hand:

npx kaloko start --scenario docs/tasks/TASK-123/qa/scenario.yml --env local
npx kaloko walk        # playwright steps; agent/manual steps: kaloko capture
npx kaloko evaluate
npx kaloko share --pr
npx kaloko feedback    # what reviewers said, with ids to answer