Asynkron TestRunner

What it does

Asynkron.TestRunner is a specialized test-running tool for .NET projects that wraps dotnet test with resilience features: hang detection, automatic isolation of problematic tests, crash/oom handling, and historical tracking of pass/fail trends. It splits test trees into branches when hangs occur, runs branches in isolation, and drills down to identify the specific failing or hanging test. It can also generate basic trend visualizations and maintain run history for regression detection.

When to use it

Use this tool when standard dotnet test is insufficient — for flaky suites, hanging tests, out-of-memory crashes, stack overflows, or when you need to isolate a problematic test in a large suite. It is expressly NOT a replacement for healthy dotnet test runs; use it only to troubleshoot instability, investigate regressions, or run in CI where timeouts and isolation are required.

What's included

• Scripts: none included in the SKILL.md snapshot (has_scripts=false)
• References: none bundled in the snapshot (has_references=false)
• Instructions: the skill provides usage patterns for running via dnx (no-install runner), common commands (list, isolate, stats, regressions), timeout and parallel flags, and guidance on how isolation and history tracking work. It documents recommended defaults (20s per-test timeout) and workflow for drilling into hanging tests.

Compatible agents

This skill fits developer-facing automation and CI orchestration agents that can run .NET SDK commands or shell out to dnx/dotnet. Agents that can manage timeouts and analyze test outputs (CI bots, build orchestrators, code-assistant agents) will benefit most.