Skill

debug-report

Systematic approach to diagnosing failing tests — trace the root cause, add instrumentation, and escalate clearly when stuck.

Invocation

How this skill is triggered — by the user, by Claude, or both

Slash command

/essential-skills:debug-report

User invocable

Model invocable

Inline context

Default effort

The summary Claude sees in its skill listing — used to decide when to auto-load this skill

When a test fails, follow this sequence before giving up:

23 lines · ~325 tokens

Stats

Parent stars0

MaintenanceGood

Last CommitFeb 9, 2026

Actions

Stats

Actions

When a test fails, follow this sequence before giving up:

Understand the test — read its name and body to know exactly which behaviour it verifies.
Walk the call stack — trace backwards from the failure to find the function responsible.
Verify inputs — confirm the failing function receives the values you expect.
Add instrumentation — insert temporary print / console.log / dbg! statements to:
- Pinpoint the branch or condition that diverges from the expected path.
- Identify the exact variable and the moment its value departs from what is correct.
Escalate if stuck — if the above steps don't lead to a fix, stop and present a structured report to the user containing:
- Test name
- Full error message and stack trace
- Inputs passed to the failing function
- Observed value vs. expected value
- Any additional observations
Link to an issue when appropriate — if the failure relates to work tracked in a GitHub issue (check the branch name for a clue), post the report as a comment on that issue. Otherwise, display it directly.