geniml

This skill should be used when working with genomic interval data (BED files) for machine learning tasks. Use for training region embeddings (Region2Vec, BEDspace), single-cell ATAC-seq analysis (scEmbed), building consensus peaks (universes), or any ML-based analysis of genomic regions. Applies to BED file collections, scATAC-seq data, chromatin accessibility datasets, and region-based genomic feature learning.

Quality

86%

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

Quality

Content

80%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 body is well-structured and highly actionable with executable examples and clean progressive disclosure to real reference files. Its main weakness is workflow clarity: batch/embedding pipelines present sequence but omit inline validation checkpoints, hitting the batch-operation cap.

Suggestions

Insert explicit validation checkpoints into the embedding and universe workflows (e.g., after tokenization: verify coverage >80% before training; after training: run evaluate_embeddings and inspect metrics before proceeding).

Tighten the Overview and 'When to Use Which Tool' sections to avoid restating the capability intros already given under Core Capabilities.

Add a brief feedback-loop note for non-convergence / low-coverage failures directly inside the relevant workflow steps rather than only in the separate Troubleshooting section.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with executable code and pointed sections, but the Overview and the 'When to Use Which Tool' bullets restate capability intros already covered, leaving minor over-explanation that could be trimmed.	4 / 5
Actionability	Provides fully executable, copy-paste-ready Python and CLI examples with real function signatures and flags covering the main Region2Vec, scEmbed, and universe-building pipelines.	5 / 5
Workflow Clarity	Numbered, clearly sequenced workflows exist, but batch operations like tokenization/universe-building and embedding training lack inline validation checkpoints (coverage verification, embedding-quality checks appear only in best practices, not as sequenced steps), capping this at 3 per the batch-operation rule.	3 / 5
Progressive Disclosure	Clear overview with well-signaled, one-level-deep references to all five existing files in references/ (region2vec, bedspace, scembed, consensus_peaks, utilities), with bulk detail appropriately offloaded and inline content limited to quick-start material.	5 / 5
	Total	17 / 20 Passed

Description

92%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 specific, trigger-rich, and clearly states both capability and use conditions for a well-scoped genomics niche. It is written in third person and avoids fluff, with only minor synonym coverage gaps in trigger terms.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions across the package's full scope — 'training region embeddings (Region2Vec, BEDspace)', 'single-cell ATAC-seq analysis (scEmbed)', 'building consensus peaks (universes)', 'ML-based analysis of genomic regions' — matching the comprehensive-coverage anchor.	5 / 5
Completeness	Explicitly answers both what (embeddings, scATAC analysis, consensus peaks, ML analysis) and when ('This skill should be used when working with genomic interval data (BED files) for machine learning tasks', 'Use for...') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Strong natural terms a genomics user would say ('BED files', 'scATAC-seq', 'chromatin accessibility', 'consensus peaks', 'universes'), but a few synonyms and the explicit '.bed' extension form are missing, so it sits just below the comprehensive anchor.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (genomic-interval ML on BED files) with distinct, specialized triggers (scATAC-seq, region embeddings, universes) and minimal overlap risk with unrelated skills.	5 / 5
	Total	19 / 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
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/ml-training/geniml/SKILL.md
Commit: 3a6c3a9

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.