CtrlK
BlogDocsLog inGet started
Tessl Logo

scientist-high

Advanced research - complex analysis and ML (Opus-tier)

44

Quality

56%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/scientist-high/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

68%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.

The content delivers two concrete, mostly-executable ML/causal-inference recipes with a clean structure and minimal padding, but it lacks explicit sequencing or validation guidance and would benefit from removing the redundant capabilities bullets. It is strongest in actionability and weakest in workflow clarity.

Suggestions

Add brief input-shape assumptions or define placeholder variables (df, features, target, treatment_conversions) so the code blocks are fully self-contained.

Either remove the 'Advanced Capabilities' bullet list since each item is demonstrated in the code, or expand it into genuinely distinct recipes not shown below.

For the ML pipeline, insert a lightweight validation step (e.g. check for NaNs/leakage before fitting, or assert CV score > baseline) to add a feedback checkpoint.

DimensionReasoningScore

Conciseness

The body is lean with two focused code blocks and a short capabilities list that assumes Claude's competence; minor redundant comments like '# Feature engineering' and a repeated persona preamble keep it just below 5.

4 / 5

Actionability

Both the sklearn grid-search pipeline and the statsmodels z-test are concrete and copy-paste ready, but they reference undefined variables (df, features, treatment_conversions), leaving a minor execution gap.

4 / 5

Workflow Clarity

Two independent recipes are presented without an explicit sequence or validation checkpoints; the [FINDING]/[STAT]/[LIMITATION] output markers help structure results but no validate-then-proceed feedback loop is shown.

3 / 5

Progressive Disclosure

No bundle files exist and the ~50-line body is organized into clear headers (Advanced Capabilities, ML Pipeline, Causal Analysis), meeting the simple-skill structure bar; it stays at 4 because the capabilities list overlaps content already shown in the code sections.

4 / 5

Total

15

/

20

Passed

Description

25%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.

The description is a terse noun phrase that names a broad domain but provides no concrete actions, no trigger guidance, and no clear niche, making it likely to overlap with other research/analysis skills. It functions more as a model-tier label than as discoverable skill guidance.

Suggestions

Rewrite in third-person action voice naming concrete tasks, e.g. 'Builds and evaluates ML models, runs causal/A-B tests, and performs time-series analysis'.

Add an explicit 'Use when...' clause with natural user phrases such as 'machine learning', 'statistical significance', or 'A/B test analysis'.

Drop internal labels like '(Opus-tier)' from the description or move them to a separate field so the description focuses on user-facing triggers.

DimensionReasoningScore

Specificity

The phrase 'complex analysis and ML' names the domain but offers no concrete actions, matching the anchor 'Names the domain but actions are minimal or generic'; it does not reach 3 because no specific task is named.

2 / 5

Completeness

The description gives only a vague 'what' ('Advanced research - complex analysis and ML') with no 'when' or 'Use when' clause, capping it at 2 per the missing-trigger-clause guidance.

2 / 5

Trigger Term Quality

'research' and 'ML' are generic keywords while 'Opus-tier' is internal jargon, and the natural phrases a user would say (e.g. 'machine learning', 'statistical analysis') are missing.

2 / 5

Distinctiveness Conflict Risk

'Advanced research' is very broad and would overlap with many analysis or research skills; the 'Opus-tier' annotation is model selection, not a distinguishing trigger.

2 / 5

Total

8

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
TurnaboutHero/oh-my-antigravity
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.