From test-engineering
Orchestrate end-to-end LowLevelTest workflows by chaining discovery, build, execution, and reporting steps
npx claudepluginhub issacchaos/local-marketplace --plugin test-engineeringThis skill uses the workspace's default tool permissions.
**Version**: 1.0.0
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
Version: 1.0.0 Category: LowLevelTests Orchestration Purpose: Chain llt-find, llt-build, llt-run, and llt-parse skills into complete test workflows
Shared Reference: See llt-common/SKILL.md for response format, data structures, validation rules, and logging instructions.
Provides automated test workflows for Unreal Engine's LowLevelTest framework. Orchestrates multiple skills to provide complete test automation with HTML reporting.
| Mode | Pipeline | Best For |
|---|---|---|
full-suite | Discover all -> build all -> run all -> parse -> report | Comprehensive validation, nightly CI |
target | Validate target -> build -> run -> parse -> report | Focused testing, iterative development |
changes | Find tests for changes -> build -> run -> parse -> report | Pre-commit validation, code review |
| Parameter | Description |
|---|---|
workflow | Workflow mode: full-suite, target, or changes |
project_root | Path to UE project root directory |
platform | Target platform (Win64, PS5, Mac, Linux, etc.) |
| Parameter | Applies To | Description |
|---|---|---|
target | target | Test target name (required for target mode) |
tags | target | Catch2 tag filter (e.g., [Core], [!Slow]) |
timeout | target | Execution timeout in seconds (default: 600) |
files | changes | List of changed source files (required for changes mode) |
include_transitive | changes | Include transitive test dependencies |
test_type | full-suite | Test type: all (default), explicit, implicit |
| Parameter | Description |
|---|---|
report_dir | Directory for HTML reports (default: ./llt-reports) |
report_name | Custom report filename |
report_only | Generate report from previous results (skip execution) |
Returns standard LLT JSON envelope (see llt-common/SKILL.md). The data field contains:
{
"workflow_type": "single-target",
"status": "success",
"total_duration_seconds": 245.67,
"tests_discovered": 1,
"tests_built": 1,
"tests_executed": 42,
"tests_passed": 40,
"tests_failed": 2,
"steps": [
{
"name": "Validate Target",
"skill": "llt-find",
"status": "completed",
"duration_seconds": 1.23,
"error": null,
"warnings": []
}
],
"report_path": "./llt-reports/llt-workflow-single-target-20260218-120000.html",
"errors": [],
"warnings": []
}
Discover Tests (invoke llt-find):
project_root, test_type, platformtotal_targets from statisticsBuild All Targets (invoke llt-build for each target):
Run All Tests (invoke llt-run for each built target):
Parse Results (invoke llt-parse for each run):
Generate HTML Report (see Report Generation below)
Validate Target (invoke llt-find):
Build Target (invoke llt-build):
action: "execute_command", the agent must execute the command via BashRun Tests (invoke llt-run):
Parse Results (invoke llt-parse):
Generate HTML Report
Find Tests for Changes (invoke llt-find-for-changes):
project_root and files listBuild Affected Tests (invoke llt-build for each discovered test):
Run Affected Tests (invoke llt-run for each built test):
Parse Results (invoke llt-parse):
Generate HTML Report
Each workflow step follows this pattern:
{name, skill, command, status: "pending"}"running", record start time"completed", store result"failed", store error messageAfter all steps complete:
status: "success" if all steps completed, "partial_success" if some failed, "failed" if critical step failedtests_discovered: count from discovery steptests_built: count from build steptests_executed, tests_passed, tests_failed: from parse step statisticstotal_duration_seconds: wall clock time from workflow start to endGenerate an HTML report with:
Report filename: llt-workflow-{workflow_type}-{YYYYMMDD-HHMMSS}.html
Default output directory: ./llt-reports/
The report uses self-contained CSS (no external dependencies). Use proper HTML escaping for all dynamic content.