init-analysis

This skill should be used when the user asks to "run initial analysis", "analyze single-cell data", "QC my data", "run bioinformatics pipeline", "generate analysis report", "explore my dataset", "do exploratory data analysis", "initial data analysis", or needs to perform quality control, dimensionality reduction, clustering, or marker analysis on single-cell biology data (CyTOF, scRNA-seq, flow cytometry, proteomics).

Quality

83%

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

82%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, token-efficient skill body that provides an executable CLI entry point, data-type-specific pipeline guidance, and genuinely non-obvious operational gotchas. The main gaps are the missing usage example for the modular approach, the lack of structured validation checkpoints, and referenced bundle files that are not present.

Suggestions

Add a short runnable example for Approach 2 (a function call on an AnnData object), since currently only the imports are shown.

Convert the 'Important Notes' into explicit validation checkpoints in the pipeline steps (e.g., verify transformation before proceeding to normalization) with fix-and-retry guidance.

Ensure the referenced files (references/*.md, scripts/run_pipeline.py, step modules) ship with the skill, or remove/inline the dead references.

Dimension	Reasoning	Score
Conciseness	Lean and efficient throughout: no concept explanations Claude already knows (PCA, UMAP, AnnData are used without preamble), and every section carries skill-specific decisions such as "Values in range [-1, 15] indicate arcsinh-transformed data" and the Leiden >50K-cell timeout note. Matches anchor 5.	5 / 5
Actionability	The CLI command is fully executable with every flag and default documented, plus a concrete output file tree. Not 5 because the modular "Approach 2" shows imports only with no usage example, leaving a minor gap in copy-paste coverage of that path.	4 / 5
Workflow Clarity	Seven steps are clearly sequenced with data-type-aware branches (CyTOF vs scRNA-seq QC and normalization), and Important Notes provide preventive checks ("Check value ranges before applying arcsinh", "Always run nan_to_num after z-score scaling"). Not 5 because these are advisory notes rather than explicit validate-then-proceed checkpoints with error-recovery loops.	4 / 5
Progressive Disclosure	Good structure: concise overview, one-level-deep references each clearly signaled with a one-line purpose (plot interpretation guide, CyTOF specifics, scRNA-seq specifics, statistical glossary). Not 5 because the referenced files (references/*.md) and the invoked scripts/run_pipeline.py and step modules are absent from the bundle, so navigation cannot be verified end-to-end.	4 / 5
	Total	17 / 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 with an explicit trigger clause and comprehensive natural-language phrasing, anchored in a well-defined single-cell biology domain. Its main weaknesses are the absence of file-extension triggers and a 'what' that is only implied through the 'when' clause rather than stated as its own capability summary.

Suggestions

State the 'what' directly in third person (e.g., "Runs a 7-step analysis pipeline... generating an HTML report") instead of embedding all capabilities in the 'when' clause.

Add file-extension triggers (.h5ad, .fcs, .csv, .h5) to match how users commonly refer to their data files.

Trim or contextualize generic phrases like "generate analysis report" and "explore my dataset" with the single-cell qualifier to reduce overlap with general analysis skills.

Dimension	Reasoning	Score
Specificity	Lists several specific actions ("perform quality control, dimensionality reduction, clustering, or marker analysis") with named data modalities (CyTOF, scRNA-seq, flow cytometry, proteomics). Not 5 because the 'what' is embedded in the 'when' clause and the concrete deliverable (7-step pipeline producing an HTML report) is never stated.	4 / 5
Completeness	Explicitly answers both 'what' ("perform quality control, dimensionality reduction, clustering, or marker analysis on single-cell biology data") and 'when' with a comprehensive list of concrete quoted trigger phrases. Clearly matches anchor 5.	5 / 5
Trigger Term Quality	Excellent natural quoted phrases ("QC my data", "explore my dataset", "do exploratory data analysis", "run initial analysis") plus modality synonyms. Falls short of anchor 5 because it omits file extensions (.h5ad, .fcs, .csv) that users often mention.	4 / 5
Distinctiveness Conflict Risk	Clear single-cell biology niche with technology-specific triggers (CyTOF, scRNA-seq, flow cytometry). Not 5 because generic phrases like "generate analysis report" and "explore my dataset" could fire for general analysis or reporting skills.	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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 5 missing	Warning

	Total	14 / 16 Passed

Repository: OpenLAIR/dr-claw
Path: skills/bioinformatics-init-analysis/skills/init-analysis/SKILL.md
Commit: d51b64e

Reviewed: 6 hours 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.