Build1 distinct publisher3 min readUpdated
Allure-Katalon Bridge fills a gap Katalon never shipped, installing by copying eight files into a project. The interesting part is stable history IDs, not the installer.
The Engineer · Build desk

Compiled by The EngineerSomething wrong?How this is made
A project called Allure-Katalon Bridge has been published on GitHub and npm under Apache-2.0, and it makes a Katalon Studio project emit Allure results by copying a handful of files into it rather than by installing a plugin [3][4]. That matters because JUnit, TestNG and Cucumber all have Allure adapters and Katalon Studio does not, which has left teams standardising on Allure dashboards with one toolchain that reports somewhere else [1].
The author describes Katalon as one of the most widely used test automation platforms, which is why the absence has been a live annoyance rather than an edge case [2]. Note that everything below comes from the project's own announcement, written by its author; none of it is independently verified here.
The delivery mechanism is the least glamorous and most defensible part. According to the announcement, the install drops a Test Listener that Katalon auto-discovers, three Groovy files under Keywords/allure covering the reporting engine, a properties reader with ALLURE_* environment overrides, and an optional keywords library, plus an allure.properties file and a categories.json tuned to Katalon and Selenium exception types [15]. Two jars come along, allure-java-commons and allure-model at 2.35.4, both Apache-2.0 from Qameta Software [16]. That is eight artifacts and no OSGi packaging, no command line requirement, and no edits to existing Test Cases or Test Suites [17][3]. Installation is a double-click installer per OS, a drag-and-drop of the project folder, a bash or PowerShell script, or npx [13]. Re-running install upgrades in place and leaves a customised allure.properties alone unless you pass --force [14].
The output claim worth reading twice is the stable history ID, which the announcement says makes retries and repeat runs appear as trend data rather than unrelated one-off results [8]. Teams that have bolted Allure onto an unsupported runner before know the failure mode: every rerun becomes a new test identity, the flake history flattens, and the trend graph becomes decoration. Around that, each test case produces one Allure result with status, timing, suite, host and thread labels and the framework named as Katalon Studio [5]; non-passed WebUI test cases get a failure screenshot, with screenshots on every case available as a one-line config change [6]; and stack traces attach automatically on failure [7]. The run ends with a self-contained allure-report HTML file with styles, scripts and data inline, so there is no local server and no allure open step [9].
Suite Collection handling is the other practical detail. Member suites combine into a single report named after the Collection instead of one report per suite, and a suite that runs more than once inside a Collection, once per browser for example, gets a separate entry rather than merging [10]. Suites that open a browser carry that browser in the name in the Suites view; pure API suites get no browser label [11]. Step-level detail, epics, severities and JSON attachments are opt-in through CustomKeywords calls from a Test Case script or Cucumber glue [12].
What to watch: the pinned 2.35.4 jars are a maintenance commitment, since Allure's report format moves and a copy-in bridge has no dependency resolver to bump them [16]. Watch whether re-running install reliably carries jar upgrades forward on projects already in the field [14], and whether Katalon responds with first-party support now that the gap has a public workaround [1].
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
A failure screenshot is captured on any non-passed WebUI test case; capturing a screenshot on every test case, pass or fail, is described as a one-line config change.
A stack trace is attached automatically on failure.
Files installed are Test Listeners/AllureTestListener.groovy (auto-discovered by Katalon, the only wiring needed), Keywords/allure/AllureReportBridge.groovy (status mapping, attachments, environment/executor/categories files), Keywords/allure/AllureConfig.groovy (allure.properties reader with ALLURE_* env var overrides), Keywords/allure/AllureKeywords.groovy (optional step and attachment/label keywords), Include/config/allure/allure.properties, and Include/config/allure/categories.json (failure categorization tuned to Katalon/Selenium exception types).
The install also adds Drivers/allure-java-commons-2.35.4.jar and Drivers/allure-model-2.35.4.jar, Apache-2.0 from Qameta Software, described as the only 2 extra jars needed, alongside a fetch-allure-jars script.
JUnit, TestNG and Cucumber each have an Allure adapter; Katalon Studio does not.
Allure-Katalon Bridge turns any Katalon Studio project into an Allure-reporting project by copying a handful of files into it, with no plugin installation, no OSGi packaging, no command line required, and no changes to existing Test Cases or Test Suites.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Single self-published account, artifacts checkable
All functional claims come from one dev.to post written by the project's own author. The claims are unusually specific and falsifiable (exact file paths, pinned jar versions 2.35.4, license, install commands), and the GitHub/npm artifacts are externally verifiable, which lifts this above pure assertion. But nothing in the cluster independently confirms runtime behaviour: no sample Allure result, no report screenshot, no third-party run, no version compatibility statement.
Published, no usage evidence
Adoption evidence stops at availability: the package is released on npm and GitHub under Apache-2.0 with documented install paths and CI templates. The cluster contains no downloads, stars, issues, named adopters or deployment reports, so there is nothing showing anyone other than the author has installed it.
Modestly overstated
Most of the post is mechanical and checkable, which keeps the gap small. It widens in three places: the unsupported framing of Katalon Studio as one of the most widely used platforms, the 'zero-touch / no side effects' absolutes offered without any test evidence, and a 'copy eight files' simplicity story that understates the shipped extras (jar fetch script, viewer launchers, CI templates, manifest file). Zero demonstrated adoption against a fairly sweeping 'fixes that' claim also pushes the gap positive.
Author promoting own project
The post is written by the project's maintainer and functions as a launch announcement, linking its own GitHub repo and npm package; the framing (competitor frameworks have adapters, Katalon does not, this fixes it) directly serves package uptake. Mitigating factors are the Apache-2.0 license, no paid tier, price or gated component in the cluster, and full disclosure of the third-party Qameta jars it depends on, so the incentive is distribution and credit rather than revenue.
Low-moderate
Confidence is limited by single-publisher, self-authored sourcing and total absence of adoption data, which caps how firmly adoption and hype-gap can be judged. It is not lower because the source is internally consistent, highly specific and unambiguous about what it claims, making the evidence and incentive readings reasonably secure.
security
The 2,500-org compromise was a Trivy problem. LiteLLM was the closing act.1 distinct publisher
build
The npm audit that works because it never installs the package1 distinct publisher
build
Your CI Build Is Slow Because The Cache Is Empty, Not Because The Base Image Is Fat1 distinct publisher
security
Shai-Hulud's fourth wave shipped with valid provenance, and that is the finding1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
dev.to
1 article · August 15, 2026