What's New With 7.0.0

March 17, 2026

TestBox 7.x series continues our mission to be the best testing framework for BoxLang and CFML. This release is focused heavily on BoxLang CLI runner enhancements, real-time streaming test execution via SSE, a powerful dry run capability, the brand-new TestBox RUN web IDE, and significant quality-of-life improvements for developers working in both BoxLang and CFML environments.

alt text

Finally, we have certified TestBox 7 for Lucee 7 and dropped support for Adobe 2021 to focus on modern engines and features.

TestBox RUN — The BoxLang Test IDE

alt text

TestBox RUN is a brand-new, BoxLang-native test runner and IDE shipped as part of TestBox 7. It is a fully-featured browser-based interface for discovering, running, and streaming your test results in real time — all powered by BoxLang and the new StreamingRunner SSE engine.

What Is It?

TestBox RUN is a self-hosted single-page web app (bx/tests/index.bxm) that you drop into your BoxLang project and open in your browser. It communicates with your existing runner.bxm endpoint and streams spec results as they complete via Server-Sent Events. No build toolchain, no external service — just BoxLang.

Feature Highlights

Real-Time Streaming Test Tree

alt text

Results appear in the UI as each spec finishes. The test tree updates live — passing specs turn green, failures turn red, and errors surface immediately with their full message — long before the suite finishes.

Beautiful, Responsive UI

alt text

alt text

• Dark / Light theme toggle with preference persistence in localStorage

• Keyboard-first UX with a full set of shortcuts:

Shortcut	Action
⌘/Ctrl + K	Focus the search bar
Esc	Clear search
⌘/Ctrl + Enter	Run all tests
⌘/Ctrl + .	Reload / rediscover tests
⌘/Ctrl + ,	Open Settings
⌘/Ctrl + B	Toggle expand/collapse all bundles
⌘/Ctrl + D	Toggle dark/light mode
⌘/Ctrl + H	Open the About / Help dialog

Live Search & Filtering

alt text

A search bar at the top lets you instantly filter the test tree by bundle, suite, or spec name. Combined with the status filter chips (Passed / Failed / Errored / Skipped), you can zero in on exactly the tests you care about.

Per-Bundle Run & Summary Strip

alt text

Each bundle card has its own ▶ Run button so you can re-run a single bundle without re-running the entire suite. A compact results strip beneath each bundle header shows pass/fail/error/skipped counts and total duration at a glance.

Debug Buffer Panel

alt text

Any output captured in the TestBox debug buffer during a run is surfaced in a collapsible per-bundle Debug Panel, accessible via a bug icon on the bundle strip. Each entry is rendered with its label and a formatted data view.

Floating Progress Widget

alt text

A floating progress widget appears during active runs, showing:

• Current bundle running

• Specs completed vs. total

• Animated progress bar with percentage

Configurable Settings

alt text

Click the gear icon (or ⌘/Ctrl + ,) to open the Settings modal, where you can configure:

• Runner URL (relative path or absolute HTTP/S)

• Directory and bundles pattern

• Recurse toggle

• Labels and excludes

All settings are saved in localStorage and applied on the next visit. Every setting can also be overridden via URL query parameters for quick CI integration:

/tests/?directory=tests.specs.integration&labels=slow&runnerUrl=/tests/runner.bxm

Getting Started with TestBox RUN

TestBox RUN is included automatically with every TestBox 7 installation under bx/tests/. Point your web server at bx/tests/index.bxm and open it in your browser.

Every ColdBox application generated with our ColdBox CLI will also include TestBox RUN out of the box.

You can also generate a new test harness with TestBox RUN support using the testbox generate command:

testbox generate harness --help

TestBox RUN requires a running web server and a runner.bxm (or runner.bx) endpoint that supports the SSE streaming protocol. If you are building a pure CLI application, use the BoxLang CLI runner with --stream instead.

Coming Soon: TestBox RUN Desktop App

Exciting news for BoxLang Desktop Runtime fans!

We are actively working on a native desktop application version of TestBox RUN, built entirely on the BoxLang Desktop Runtime. The desktop app will connect to any web-accessible runner URL — local or remote — and provide the same beautiful real-time streaming UI without a browser. Stay tuned at testbox.run for announcements, early access, and previews.

Engine Support

Adobe 2021 has been dropped and Lucee 7 is now fully supported and certified.

Engine	Supported
BoxLang 1.x+	✅ PREFERRED
Lucee 5.x	✅ DEPRECATED
Lucee 6.x	✅
Lucee 7.x	✅ NEW
Adobe 2023	✅ DEPRECATED
Adobe 2025	✅
Adobe 2021	❌ Dropped

Adobe 2021 is no longer supported as of TestBox 7. Please upgrade to Adobe 2023+ or migrate to BoxLang.

TestBox RUN

Dry Run & Spec Discovery

Two long-requested features have finally landed: the ability to discover which specs will run without executing them, and a full dry run mode. These are invaluable for large test suites where you need to audit coverage or validate filtering before committing to a full run.

Call the dryRun() method on your TestBox instance to execute the full discovery and filtering pipeline and get a complete report of what would have run — without actually invoking any spec closures.

var tb      = new testbox.system.TestBox( bundles = "tests.specs" );
var results = tb.dryRun();

The BoxLang CLI runner also gains a --dry-run flag that performs the same discovery and filtering process, outputting a structured report of all suites and specs that would have been executed:

alt text

./testbox/bin/run --dry-run

The output lists every suite and spec that would be executed, along with their labels and any skip reasons — perfect for CI pipeline auditing and test inventory reporting.

Dry Run JSON Output

Pass --dry-run=json to receive the full discovery payload as raw JSON instead of the formatted text output. This is ideal for programmatic consumption — CI tools, custom reporters, or any external script that needs to process the test inventory:

./testbox/bin/run --dry-run=json
./testbox/bin/run --dry-run=json --bundles=tests.specs.MySpec | jq .

Dry run mode respects all the same filtering options as a normal run: --labels, --bundles, --directory, --testSuites, --testSpecs, etc.

Streaming Runner & Real-Time Output (TESTBOX-441, TESTBOX-442)

TestBox 7 introduces a brand-new StreamingRunner that pushes test results to the client as they complete via Server-Sent Events (SSE). This means you no longer have to wait for the entire suite to finish before seeing results — each spec result is streamed in real time. This of course, requires an SSE-capable client, but it opens up powerful new workflows for CI pipelines and large test suites where immediate feedback is crucial.

StreamingRunner

The new StreamingRunner.cfc is available under testbox.system.runners.StreamingRunner and can be used programmatically or wired into any SSE-capable endpoint.

component {
    function streamTests( event, rc, prc ) {
        event.setHTTPHeader( name="Content-Type", value="text/event-stream" );
        event.setHTTPHeader( name="Cache-Control", value="no-cache" );

        new testbox.system.runners.StreamingRunner(
            bundles  = "tests.specs",
            options  = {},
            reporter = "text"
        ).run();
    }
}

BoxLang CLI --stream Flag

The BoxLang CLI runner gains a --stream flag that activates real-time output of test results as they execute rather than buffering until completion:

./testbox/bin/run --stream
./testbox/bin/run --directory=tests.specs --stream

Streaming output is especially useful in CI environments where you want live progress rather than waiting for the full suite to complete before seeing any output.

BoxLang Runner Enhancements

The BoxLang CLI runner has received a significant set of new options that give you fine-grained control over output, failure reporting, and performance analysis.

--show-failed-only (TESTBOX-444)

Focus your terminal output exclusively on failures and errors, suppressing all passing and skipped specs:

./testbox/bin/run --show-failed-only

--stacktrace Control (TESTBOX-445)

Choose how much stack trace detail is shown for failures and exceptions. The default is short to keep output readable:

./testbox/bin/run --stacktrace=short   # default: condensed first frame
./testbox/bin/run --stacktrace=full    # complete Java/BoxLang stack trace

Output & Performance Options (TESTBOX-446)

A full suite of output-control and performance-analysis flags are now available:

# Show or hide passing specs (default: true)
./testbox/bin/run --show-passed=false

# Show or hide skipped specs (default: true)
./testbox/bin/run --show-skipped=false

# Abort the run after N failures
./testbox/bin/run --max-failures=10

# Flag any spec that takes longer than N milliseconds as slow
./testbox/bin/run --slow-threshold-ms=500

# Print a summary of the N slowest specs at the end of the run
./testbox/bin/run --top-slowest=5

Combining these gives you a clean, focused workflow:

./testbox/bin/run --show-failed-only --stacktrace=short --max-failures=5 --top-slowest=3

Application Mappings Auto-Load (TESTBOX-440)

The BoxLang runner now automatically attempts to load Application.bx mappings from your project root before executing tests. This means any custom path mappings, datasources, or settings defined in your Application.bx are available to your specs without any extra configuration.

This brings the BoxLang CLI runner experience much closer to running tests in a full web server environment, reducing the friction of maintaining separate CLI-specific configuration.

New ConsoleUtil Toolkit (TESTBOX-443)

A new ConsoleUtil class has been added to testbox.system.util.ConsoleUtil to provide a rich set of helpers for building beautiful CLI tooling on top of TestBox. It includes support for:

• ANSI color and style output helpers

• Bordered and padded boxes for drawing section headers

• Progress indicators and spinner helpers

• Table rendering for structured output

• Symbols and icons normalized across platforms

var console = new testbox.system.util.ConsoleUtil();

console.printLine( console.green( "All tests passed!" ) );
console.printBox( "Suite: My Application Tests", style="double" );
console.printKeyValue( "Total Specs", 142 );
console.printKeyValue( "Failures",    0,  color="green" );

ConsoleReporter — Hide Skipped Tests (TESTBOX-433)

The ConsoleReporter now accepts a hideSkipped option (default false) so you can suppress skipped spec output for cleaner terminal readability when you have many pending specs:

var testbox = new testbox.system.TestBox(
    bundles  = "tests.specs",
    reporter = {
        type        : "testbox.system.reports.ConsoleReporter",
        options     : { hideSkipped : true }
    }
);

From the BoxLang CLI runner, use --show-skipped=false (see above).

Suite Filtering Improvements (TESTBOX-435)

TestBox now performs direct suite name matching when the testSuites filter is active. Previously, nested suites could be skipped even when they exactly matched the filter value. Now, if a suite's name is a direct match for the filter, it will always be included regardless of nesting depth.

This makes granular suite targeting far more reliable from all runners:

# Via CLI runner
./testbox/bin/run --testSuites="My Integration Suite"

# Via TestBox directly
new testbox.system.TestBox(
    bundles    = "tests.specs",
    testSuites = "My Integration Suite"
).run();

TestBox CLI Updates

alt text

The testbox-cli CommandBox module has received a major update in version 1.8.0 to align with TestBox 7.

New Commands

testbox info

A new testbox info command displays detailed information about the current TestBox installation, including the installed version, installation path, and key configuration detected in your project:

testbox info

testbox reinstall

A new testbox reinstall command forces a clean reinstallation of the TestBox CLI package itself — useful when you suspect your current installation is corrupted or outdated:

testbox reinstall

This command reinstalls the testbox-cli CommandBox module only, not the TestBox library itself. To reinstall the TestBox library, use box install testbox --force.

Streaming Results via testbox run --streaming

The testbox run command now supports a --streaming flag that taps into the new StreamingRunner to deliver real-time test results to your terminal as each spec completes rather than buffering until the entire suite finishes:

testbox run --streaming

Pair it with --verbose to include all passing specs in the live output — by default only failures and skipped specs are persisted:

testbox run --streaming --verbose

Streaming output is ideal for CI environments where you want immediate feedback on test progress rather than waiting for the full suite to complete before seeing any results. It works best with BoxLang, but can be used with CFML engines that support SSE as well.

TestBox RUN Integration

The CLI now ships with support for the new TestBox RUN runner for BoxLang, enabling native SSE-based streaming runs directly from the CommandBox CLI without needing a separate web server.

Template Updates

Both CFML and BoxLang generation templates have been refreshed to reflect the latest TestBox 7 conventions and best practices.

Bug Fixes

• getTestBoxDescriptor() argument was previously ignored and is now correctly applied.

• The testbox package is now used as the single source of truth for version and descriptor resolution rather than duplicating logic across the CLI module, preventing potential drift.

Improved Help Messages

All commands have received updated help text with clearer descriptions, usage examples, and argument documentation.

Release Notes

Bugs

#	Title
TESTBOX-432	toBeWithCase needs the caseSensitive flag set
TESTBOX-436	Trim is killing leading ANSI escapes by removing chr(27) in the console reporter
TESTBOX-437	Update location detection of TestBox library on Windows for BoxLangRunner.bx
TESTBOX-438	CLI args not being passed to runner on Windows using BoxLang
TESTBOX-439	Fix TestBox scanning from the root if directory was not passed but bundles were using the BoxLang Runner

Improvements

#	Title
TESTBOX-430	BoxLang PRIME updates for detecting CGI scopes when doing CLI testing instead of web
TESTBOX-434	Improve static block usage on BoxLang runner
TESTBOX-435	Determine if a suite is a DIRECT match for the testSuites filter; if so, run it — allows reliable suite filtering via runners
TESTBOX-440	Attempt to load Application mappings using the BoxLangRunner

New Features

#	Title
TESTBOX-213	Allow retrieving specs that will be run
TESTBOX-324	Enable a "Dry Run" so you can get a list of tests without actually running them
TESTBOX-431	Lucee 7 certification
TESTBOX-433	ConsoleReporter option hideSkipped to omit skipped tests from output
TESTBOX-441	New StreamingRunner to provide execution streaming of tests via SSE
TESTBOX-442	New --stream option for the BoxLang runner to stream results and tests in real-time
TESTBOX-443	New ConsoleUtil for building beautiful CLI tooling
TESTBOX-444	New --show-failed-only option for the BoxLang Runner
TESTBOX-445	New --stacktrace=short|full option for the BoxLang runner
TESTBOX-446	New BoxLang runner options: --show-passed, --show-skipped, --max-failures, --slow-threshold-ms, --top-slowest

Tasks

#	Title
TESTBOX-447	Drop Adobe 2021 support