Skip to content
CheckcraftDownload
API and log checks without code

A successful request can still return the wrong result

Check response fields, XML and the matching logs. Save the expected result once and repeat the same checks after a change.

Get Checkcraft FreeSee a working example

Windows and Linux · Free with no time limit · No payment card

Download directly, no account or email needed. Bought Pro? Sign in

Get help with your first test →

A real result from Checkcraft
HTTP 200, but the order is not paid, itemCount is a string, debug is unexpected and total is 2500 instead of the calculated 2000. Passed checks remain visible alongside the errors.
HTTP 200, but the order is not paid, itemCount is a string, debug is unexpected and total is 2500 instead of the calculated 2000. Passed checks remain visible alongside the errors.
See the rules and repeat this result →

See what a check finds

This response came back with HTTP 200. Press the button and see what the rules catch: this is real logic running in your browser.

What would you like to check?

Each example includes sample data and an explanation of the result. You can explore them before downloading.

Check without installingYour own JSON, right in the browserRun your first testSample data and expected resultsCheck every array item1,000 objects without manual inspectionFind unexpected JSON fieldsRequired fields and allowed propertiesConnect responses and logsTrace a workflow using traceIdValidate XMLValues, attributes and repeated nodesCreate a draft with AIFrom documentation and sample data
From response to a repeatable check

See what failed and why

A status code tells you whether the server answered. Your saved rules tell you whether the answer matches the requirement.

01

Start from data you already have

Use a cURL request, an OpenAPI description or a JSON example. The first sample works with synthetic data, so you can inspect the result before connecting your services.

Import a request →
02

Make the expectation explicit

Choose required fields, allowed keys and expected values. Check every array item or calculate the total that the service should return.

Check 1000 array items →
03

Follow the operation into its logs

Use the trace identifier to find the related records. Extract embedded JSON and keep the log rules alongside your request checks.

Connect API and log checks →

See the current application

Current Checkcraft screenshots using sample data. Select a task, inspect the configuration and open the walkthrough.

A cURL request in the editor

Method, URL and JSON were imported into request fields. This sample uses localhost; no external service request was sent.

A cURL request in the editor

Follow the worked example →Click the screenshot for full size

Review the steps before adding them

Checkcraft AI prepared a draft from JSON. This example used structured parsing without calling a language model. Review and edit the expectations before use.

Review the steps before adding them

Follow the worked example →Click the screenshot for full size

One rule for every item

The loop applies nested rules to every array item. The sample contains 1,000 objects with an error in the last one.

One rule for every item

Follow the worked example →Click the screenshot for full size

From a log record to a JSON check

Select a record by traceId, extract JSON from message and check its fields. The screenshot uses a local sample log record.

From a log record to a JSON check

Follow the worked example →Click the screenshot for full size

XML rules in the editor

Expected values are defined in individual steps. The walkthrough includes XML samples and a test case.

XML rules in the editor

Follow the worked example →Click the screenshot for full size

Configure a local response

Prepare a route and response body for your check. This is the configuration screen for a sample mock server.

Configure a local response

Follow the worked example →Click the screenshot for full size

Prepare a runner command in the app

Select checks and prepare the runner command. The screenshot shows configuration, not a completed Jenkins run.

Prepare a runner command in the app

Follow the worked example →Click the screenshot for full size

One small experiment

Watch the same check pass and fail

The sample expects status to be OK. Run the valid input, then change it to ERROR and run the unchanged rule. You can see the exact mismatch before spending time on a real integration.

Open the sample files

Sample files are available without registration. Running the desktop app requires a verified account.

Passing sample
[OK] status = 'OK' -> = 'OK'
Failing sample
[FAIL] status = 'ERROR' -> (expected 'OK')

Output from two actual local runs with the same test case.

Find your task

Work through a concrete example

Practical guide

How to create API tests without code

You will import a test case, run it against two inputs and see the exact field that differs.

Open guide →Practical guide

Validate every element of a JSON array

The downloadable example contains 1000 items. The broken input puts a defect at the end so you can verify that the test reaches it.

Open guide →Practical guide

Create API test drafts with local AI

You can review proposed test cases before adding them, then run the saved rules on passing and deliberately failing data.

Open guide →Practical guide

Automate API and log checks with Kibana

The request identifier selects the operation to inspect, and saved rules validate fields from the matching log data.

Open guide →Practical guide

Create XML response tests without code

The example validates XML content and a tag whitelist. Its broken input makes a saved rule fail.

Open guide →Practical guide

Test API totals, discounts and calculations

The example sums item amounts, stores the expected total in a variable and compares it with the returned total.

Open guide →

Browse all 20 guides →

Choose a workflow that fits

Where Checkcraft can help

If you already use Postman

Postman can send requests and run scripted tests. Checkcraft gives you an editable step-based workflow for response and log checks. Compare the same task in both tools before changing your process.

Compare the workflow →

If you already write tests in code

Code gives you flexibility and integration with your existing test suite. Checkcraft can help when teammates need to inspect and change a repeatable check through an interface.

See the CI workflow →

If you want AI to prepare the draft

Local AI tools can help turn clear examples into editable rules. You still decide the expected behaviour and check the draft. For structured inputs, the parser can create steps without calling the language model.

See an actual generated draft →

Know what you are installing

Checkcraft is a desktop application for Windows 10/11 x64 and Linux x64. The archive includes a runtime and local AI components. Your account shows the current download size.

Download and first-run instructions · Security and data handling

Start with the free edition

Free has no time limit. Some saved objects, sequences and CI suites have limits. The edition page explains those limits before you choose a workflow.

Read the Free limits

Try your own check

See the result on a small example first

Start with the sample, inspect the failure and then try a sanitized example from your own API. Free is available for Windows and Linux with no time limit.

Download FreeDownload and first run

No sign-up or email needed, the archive is ready in a couple of minutes. See the edition limits.

Ask a person

Get help with your first test

Tell us what you want to test or where you got stuck. We can help you start Checkcraft, choose an example and understand your first run. You can ask before downloading.

Ask in TelegramSend an email

The Telegram group is public: do not post passwords, tokens or confidential work data. A description of your task is enough to start.

You can also email us support@checkcraft.ru

No account required. We will reply by email. This is a support request, not a newsletter subscription.

See exactly what the check found

These failures come from actual test-case runs. The detailed report places rules next to the checked values. Click the screenshot to inspect the full screen.

HTTP 200 hides four different problems

HTTP 200, but the order is not paid, itemCount is a string, debug is unexpected and total is 2500 instead of the calculated 2000. Passed checks remain visible alongside the errors.

HTTP 200, but the order is not paid, itemCount is a string, debug is unexpected and total is 2500 instead of the calculated 2000. Passed checks remain visible alongside the errors.

Download this report’s test case →Click the screenshot for full size

Sample responses with errors · corrected

The error is in items[999]

Actual detailed HTTP run report with rules next to the checked values. Open the screenshot to see the full run screen. The report is scrolled to items[999], the failing item out of 1,000.

Actual detailed HTTP run report with rules next to the checked values. Open the screenshot to see the full run screen. The report is scrolled to items[999], the failing item out of 1,000.

Follow the worked example →Click the screenshot for full size

A log value did not match

The record is selected by traceId. Its embedded message JSON contains FAILED instead of PAID, while the amount check passes. Actual detailed Log run report with rules next to the checked values. Open the screenshot to see the full run screen.

The record is selected by traceId. Its embedded message JSON contains FAILED instead of PAID, while the amount check passes. Actual detailed Log run report with rules next to the checked values. Open the screenshot to see the full run screen.

Follow the worked example →Click the screenshot for full size

XML does not match the expectation

The XML contains ERROR instead of OK and an unexpected debug tag. The amount check passed. Actual detailed HTTP run report with rules next to the checked values. Open the screenshot to see the full run screen.

The XML contains ERROR instead of OK and an unexpected debug tag. The amount check passed. Actual detailed HTTP run report with rules next to the checked values. Open the screenshot to see the full run screen.

Follow the worked example →Click the screenshot for full size

A calculation exposed a mismatch

The items add up to 300, but total is 301. The formula appears next to the mismatch. Actual detailed HTTP run report with rules next to the checked values. Open the screenshot to see the full run screen.

The items add up to 300, but total is 301. The formula appears next to the mismatch. Actual detailed HTTP run report with rules next to the checked values. Open the screenshot to see the full run screen.

Follow the worked example →Click the screenshot for full size