CtrlK
BlogDocsLog inGet started
Tessl Logo

jam-workspace/reproduce-first-debugging

Evidence-gated debugging discipline for coding agents: no edits before a captured reproduction, red test before fix, bisect over guesswork, root cause over symptom patch

74

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Overview
Quality
Evals
Security
Files

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A high-quality, highly actionable debugging methodology with executable commands, explicit validation gates, and feedback loops throughout. The only marginal area is conciseness, where a few motivational asides could be trimmed.

DimensionReasoningScore

Conciseness

Dense and command-laden with no padding about concepts Claude already knows, but a few motivational sentences ("Guess-and-check past three loops is how sessions burn hours", "A minimal repro often names the root cause by itself") could be trimmed.

4 / 5

Actionability

Fully executable guidance: copy-paste git bisect block, report template, and concrete commands across pytest/jest/go test/curl/CLI plus a scripted pdb one-liner; the common cases are covered.

5 / 5

Workflow Clarity

A clearly sequenced 7-step pipeline with mandatory ordering ("red test → fix → green"), explicit validation checkpoints (Step 6.3 repro re-run, 6.4 baseline comparison), and feedback loops (3-hypothesis cap → bisect, escape hatch after 3 strategies).

5 / 5

Progressive Disclosure

Well-organized single-file structure with clear section headers (Iron Rule, Steps 1–7, Judgment rules) and no nested references; at ~120 lines with all content inline it is cohesive, though a couple of integral blocks (report template, patch smells) could optionally live in references.

4 / 5

Total

18

/

20

Passed

Description

95%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, trigger-rich description with excellent natural keywords and exemplary when-guidance including negative exclusions. Its only weakness is that the distinctive reproduce-first methodology is implied by the name rather than stated in the description itself.

DimensionReasoningScore

Specificity

Names several concrete actions ("debug, troubleshoot, diagnose, investigate, or root-cause a failure") and concrete failure types (bug, error, crash, flaky test, regression), but never states the distinctive reproduce-first mechanism, leaving a minor coverage gap.

4 / 5

Completeness

Explicitly answers both what (debug/root-cause failures) and when ("Load whenever the task is to debug…", "Load BEFORE reading the suspect code"), with concrete trigger phrases and explicit negative exclusions for TDD-red and explain-only cases.

5 / 5

Trigger Term Quality

Comprehensive natural terms with synonyms (debug/troubleshoot/diagnose/investigate/root-cause) and concrete user phrases like "bug", "crash", "flaky test", and "it doesn't work" report.

5 / 5

Distinctiveness Conflict Risk

Clear failure/root-cause debugging niche with distinct triggers; the explicit exclusions (expected-red TDD, explain-only questions) actively reduce conflict with related skills.

5 / 5

Total

19

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Reviewed

Table of Contents