analyze-results

Analyze ML experiment results, compute statistics, generate comparison tables and insights. Use when user says "analyze results", "compare", or needs to interpret experimental data.

Quality

74%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/analyze-results/SKILL.md

Quality

Content

65%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 well-organized, concise instruction-only workflow with a strong findings framework, but it stays at a fairly abstract level without executable specifics or validation checkpoints for batch result processing.

Suggestions

Add concrete execution details: name the tool/library for statistics and table generation, and show a minimal example command or snippet for parsing JSON/CSV results.

Insert a validation checkpoint early in the workflow (e.g., verify result files exist and parsed data is non-empty before computing statistics) to support batch operation safety.

Specify how 'delta vs baseline' and 'mean +/- std' should be computed/formatted so the analysis is reproducible.

Dimension	Reasoning	Score
Conciseness	The body is lean, bullet-driven, and assumes Claude's competence without explaining concepts Claude already knows; not 5 because a few sections (Step 5, Output Format) could be trimmed further, and not 3 because there is no noticeable padding.	4 / 5
Actionability	Provides a concrete reusable framework (Observation/Interpretation/Implication/Next step, delta-vs-baseline, mean+/-std) but lacks the specific executable details (libraries, commands, how statistics are computed) needed to fully execute; matches the 'some concrete guidance but incomplete; missing key details' anchor.	3 / 5
Workflow Clarity	Steps 1-5 are clearly sequenced, but processing multiple result files is a batch operation with no validation/verification checkpoint (e.g., confirm files exist, sanity-check parsed data), so per the rubric cap workflow clarity cannot exceed 3.	3 / 5
Progressive Disclosure	No bundle files exist and the body is under 50 lines with well-organized sections (Workflow, Steps 1-5, Output Format), meeting the simple-skill exception that allows a 5 with just clear section organization.	5 / 5
	Total	15 / 20 Passed

Description

83%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 clearly states both capability and explicit trigger conditions with natural user phrases. Minor gaps in synonym/file-extension coverage and a slightly generic 'compare' trigger keep it just below the top tier.

Suggestions

Add common synonyms and file extensions to triggers (e.g., 'evaluate', 'benchmark', 'sweep', '.json/.csv results') for broader natural-term coverage.

Tighten the generic 'compare' trigger with domain context (e.g., 'compare runs or models') to reduce overlap risk with general comparison skills.

Dimension	Reasoning	Score
Specificity	Names the domain ('ML experiment results') and three concrete actions ('compute statistics, generate comparison tables and insights'), matching the 'lists several specific actions; minor gaps' anchor; not 5 because coverage is not exhaustive and not 3 because more than 1-2 actions are named.	4 / 5
Completeness	Explicitly states what it does and provides a concrete 'Use when...' trigger clause with quoted phrases, matching the 'clearly and explicitly answers both what AND when' anchor.	5 / 5
Trigger Term Quality	Includes natural user phrases ('analyze results', 'compare', 'interpret experimental data') giving good keyword coverage; not 5 because common synonyms (e.g., 'evaluate', 'benchmark') and file extensions are missing, and not 3 because more than 'some relevant keywords' are present.	4 / 5
Distinctiveness Conflict Risk	The ML experiment-results niche is mostly distinct with clear triggers; not 5 because the bare term 'compare' is generic enough to risk minor overlap with related skills, and not 3 because the ML framing keeps it largely distinguishable.	4 / 5
	Total	17 / 20 Passed

Validation

87%

Warnings & errors only

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

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/analyze-results/SKILL.md
Commit: f4f20f9

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.