deeptools
NGS analysis toolkit. BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), for ChIP-seq, RNA-seq, ATAC-seq visualization.
74
66%
Does it follow best practices?
Impact
90%
1.50xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./scientific-skills/deeptools/SKILL.mdQuality
Discovery
82%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong, domain-specific description that excels at listing concrete capabilities and using natural trigger terms from the bioinformatics/NGS domain. Its main weakness is the absence of an explicit 'Use when...' clause, which would help Claude know precisely when to select this skill. The dense, abbreviation-heavy style works well for the target audience but could benefit from a brief trigger statement.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user needs to process BAM files, generate bigWig tracks, perform deepTools-style QC, or create heatmaps/profiles for sequencing data.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), and names specific assay types (ChIP-seq, RNA-seq, ATAC-seq). | 3 / 3 |
Completeness | Clearly answers 'what does this do' with specific actions and data types, but lacks an explicit 'Use when...' clause or equivalent trigger guidance, which caps this at 2 per the rubric guidelines. | 2 / 3 |
Trigger Term Quality | Excellent coverage of natural terms a bioinformatics user would say: NGS, BAM, bigWig, QC, correlation, PCA, fingerprints, heatmaps, profiles, TSS, peaks, ChIP-seq, RNA-seq, ATAC-seq. These are precisely the terms users in this domain would use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche in NGS/deepTools-style analysis with very specific domain terminology (BAM, bigWig, ChIP-seq, ATAC-seq) that is unlikely to conflict with other skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides excellent actionable content with concrete, executable commands and good coverage of multiple NGS analysis types. However, it is severely bloated with unnecessary meta-instructions (how to handle users, example interactions, key reminders) and content that should either be in reference files or omitted entirely. The verbosity significantly undermines its effectiveness as a skill file.
Suggestions
Remove the 'Handling User Requests', 'Example Interactions', and 'Key Reminders' sections entirely—Claude doesn't need instructions on how to interact with users or reminders to document commands.
Move the detailed 'Best Practices', 'Effective Genome Sizes' table, and 'Reference Documentation' descriptions to reference files, keeping only a brief pointer in the main skill.
Add explicit validation checkpoints within each workflow (e.g., 'verify output file exists and is non-empty' after each major step) rather than only validating inputs upfront.
Consolidate the 'When to Use This Skill' section into the YAML frontmatter or remove it—Claude can determine applicability from the skill's content.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~400+ lines. Contains extensive sections explaining when to use the skill, how to handle user requests, example interactions, and key reminders that are unnecessary for Claude. Sections like 'Handling User Requests' and 'Example Interactions' teach Claude how to be a chatbot, which it already knows. The 'Reference Documentation' section redundantly describes what each reference file contains with 'When to use' guidance that Claude can infer. | 1 / 3 |
Actionability | Provides fully executable, copy-paste ready commands throughout (bamCoverage, bamCompare, computeMatrix, plotHeatmap, etc.) with specific parameters. Includes concrete validation scripts, workflow generator commands, and troubleshooting commands like `samtools index`. | 3 / 3 |
Workflow Clarity | Workflows are listed with numbered steps and the general pattern (QC → Normalization → Comparison/Visualization) is clear. However, validation checkpoints within the actual analysis workflows are mostly implicit—the skill says 'validate files first' but doesn't embed explicit validation/verification steps within each workflow sequence. The workflows largely delegate to external references and generated scripts rather than showing complete sequences with feedback loops. | 2 / 3 |
Progressive Disclosure | References to external files (references/tools_reference.md, references/workflows.md, etc.) are well-signaled and one level deep. However, the SKILL.md itself is a monolithic wall of text with too much inline content that should be in reference files—the 'Handling User Requests', 'Example Interactions', effective genome sizes table, and detailed best practices sections bloat the main file significantly. | 2 / 3 |
Total | 8 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (530 lines); consider splitting into references/ and linking | Warning |
metadata_version | 'metadata.version' is missing | Warning |
Total | 9 / 11 Passed | |
- Repository
- K-Dense-AI/claude-scientific-skills
- Commit
b58ad7e
- 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.