CtrlK
BlogDocsLog inGet started
Tessl Logo

investigate-first

Diagnose ambiguous failures before editing. Use for unknown causes, intermittent behavior, performance regressions, or investigations needing evidence-ranked hypotheses.

68

Quality

82%

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

SKILL.md
Quality
Evals
Security

Quality

Content

86%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.

An exceptionally lean, well-structured investigation methodology with explicit gating checkpoints and no token waste. Its only gap is the absence of a worked example or error-recovery loop to push actionability and workflow clarity to full marks.

Suggestions

Add one short worked example (e.g., symptom -> top hypothesis -> falsifying check -> confirmed cause) to make the ranking method concrete and lift actionability.

Add an explicit error-recovery step ('if no hypothesis survives falsification, widen the trace or restate the symptom') to give workflow clarity a feedback loop.

Optionally include a one-line stopping criterion format for the reported proof so the output is unambiguous.

DimensionReasoningScore

Conciseness

Every line is actionable guidance with no padding and no explanation of concepts Claude already knows, matching the 'lean and efficient; every token earns its place' anchor.

5 / 5

Actionability

Concrete, specific instruction ('Trace inputs, state transitions, ownership boundaries, and failure output'; 'Rank hypotheses by evidence and cheap falsification value') with only minor gaps and no worked example, fitting 'mostly executable guidance; minor gaps' rather than fully copy-paste-ready 5.

4 / 5

Workflow Clarity

A clear sequence with explicit gating checkpoints ('Do not edit until one credible mechanism explains evidence'; 'Stop exploration when evidence is sufficient'), but no error-recovery feedback loop, matching 'clear sequence with most checkpoints present; minor validation gaps'.

4 / 5

Progressive Disclosure

Under 50 lines, no external references needed, and well-organized (title, framing sentence, bulleted method, closing instruction), so the simple-skill exception awards 5 for clean structure.

5 / 5

Total

18

/

20

Passed

Description

78%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 description that pairs a concrete 'what' with an explicit 'Use for' trigger list of natural phrases. It is held back only by listing a single primary action and by slightly limited synonym coverage.

Suggestions

Add 1-2 more concrete actions (e.g., 'rank hypotheses by evidence, isolate root cause') to lift specificity toward several distinct actions.

Include common synonyms users might say, such as 'flaky tests' or 'mystery bugs', to broaden trigger term coverage.

Consider naming the output artifact (e.g., 'produce an evidence-ranked hypothesis report') to sharpen distinctiveness from generic debugging skills.

DimensionReasoningScore

Specificity

Names the domain ('ambiguous failures') and a concrete action ('Diagnose', 'evidence-ranked hypotheses') but does not list several specific actions, matching the '1-2 concrete actions, not comprehensive' anchor.

3 / 5

Completeness

Clearly states what ('Diagnose ambiguous failures before editing') and an explicit 'Use for ...' when-clause with concrete trigger phrases, matching the 'clearly and explicitly answers both what AND when' anchor.

5 / 5

Trigger Term Quality

'unknown causes, intermittent behavior, performance regressions' are natural user phrases with good coverage, but a few synonyms are missing, fitting 'good keyword coverage; a few natural terms missing' rather than the comprehensive 5.

4 / 5

Distinctiveness Conflict Risk

A clear investigation niche with distinct triggers ('evidence-ranked hypotheses') gives mostly distinct positioning with only minor overlap risk against general debugging skills, so 4 rather than 5.

4 / 5

Total

16

/

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.

Repository
JuliusBrussee/caveman
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.