skills/straub/agent-skills/running-tests

running-tests

SKILL.md

Running Tests

Test suites can sometimes take a very long time to run, consume lots of system resources, and generate lots of output. To save time and preserve your context, follow this simple guide.

When to Use This Skill

āœ… Use if:

  • Running any test command (npm test, node --test, pytest, etc.)
  • Expecting more than 5 lines of output
  • Want to preserve context tokens for further analysis

āŒ Don't use if:

  • Running a single quick test file (< 1 second)
  • Already redirecting output manually

Guidelines

  • Always redirect test output to a file!
npm test > /tmp/test-output.txt 2>&1
  • Analyze that file as needed after the run is finished
tail -n20 /tmp/test-output.txt

wc -l /tmp/test-output.txt

grep 'fail' /tmp/test-output.txt

Default Pattern

Every test run should follow this pattern:

npm test > /tmp/test-output.txt 2>&1
tail -20 /tmp/test-output.txt
Weekly Installs
4
First Seen
Feb 4, 2026
Installed on
opencode4
claude-code4
github-copilot4
gemini-cli3
replit3
codex3