agentsclimarketplace

Test failure triage

Skill charlieviettq/awesome-agent-skill/.claude/skills/test-failure-triage

Curated skill pack for LLM agents in engineer and science workflow (Cursor & Claude ready).

Install
npx -y skills add charlieviettq/awesome-agent-skill --skill test-failure-triage

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 22 stars22 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Systematically triage failing tests by grouping errors, fixing root causes in priority order, and verifying incrementally. Use when CI fails, test suite is broken, or multiple tests fail after a refactor or merge.

SKILL.md

1.5 KB, 282 tokens by cl100k_base, as published. Nobody here has run it

Test failure triage

Workflow

1. Baseline run

Run the project's standard test command and capture full output (counts, error types, files).

2. Group failures

Group by:

  • Error type: ImportError, AttributeError, AssertionError, etc.
  • Root file/module: Same file causing multiple failures.
  • Likely cause: Recent change area (use git diff, recent commits).

Prioritize groups with highest downstream impact (imports/config before assertion tweaks).

3. Fix order

  1. Infrastructure: imports, deps, config, env
  2. API/signature changes: renamed modules, function args
  3. Logic: assertion and business-logic failures

4. Verify per group

After each group fix, re-run a focused subset (file or pattern), then full suite at the end.

# Example patterns (adapt to project)
pytest path/to/test_file.py -v
pytest -k "pattern" -v
make test

Practices

  • Fix one group at a time; do not mix unrelated fixes in one commit if avoidable.
  • Prefer minimal changes; match project test style.
  • Document external factors (flaky tests, env-only failures).

Output

Short summary: total failures, groups fixed, remaining risks, final full-suite result.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.