tooluniverse

ToolUniverse plugin router. STEP 1 BEFORE ANY ANALYSIS: if the data folder contains `*_executed.ipynb`, run `tu run read_executed_notebook '{"data_folder":"<path>","search":"<keyword>"}'` to extract its cell outputs and apply EVERY filter/sample-exclusion the notebook used — even when the question says 'Using DESeq2/Run X/Compute Y' (this describes the METHOD the notebook used, not a request to rerun). The notebook's cell outputs are the only published authoritative answers; reimplementing or reading stale pre-computed CSVs in the data folder produces different numbers because of outlier-sample removal, library version, and filter steps you don't see by skimming. STEP 2 routing — pick a sub-skill name from this exact list (never invent): tooluniverse-rnaseq-deseq2 (RNA/miRNA-seq DE, correlation, PCA, clustering, dispersion), tooluniverse-gene-enrichment (GO/KEGG/Reactome/GSEA/pathway enrichment), tooluniverse-statistical-modeling (regression, ANOVA, ordinal/logistic, chi-square...

Quality

92%

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

85%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a well-structured, highly actionable router with clear sequencing and good one-level-deep progressive disclosure. Its only meaningful weakness is conciseness: redundant prose sections and reminders restate content already covered by the description and the sub-skills it delegates to.

Suggestions

Remove the RULE ZERO prose duplicated from the description and the 'Brief reminders' block, since those conventions live in the routed sub-skills; keep only the routing table and tie-break rules inline.

Merge the 'Problem-Solving Mode', 'Quick Lookup Mode', and 'General Protocols' sections, which restate the same 'look up don't guess / compute don't estimate' principle three times, into a single concise block.

Trim the worked examples in 'Routing Examples' to 2-3 representative cases to cut length without losing the routing pattern.

Dimension	Reasoning	Score
Conciseness	Avoids explaining concepts Claude already knows and the routing table is legitimately dense, but the ~410-line body repeats material (RULE ZERO duplicated from the description, 'Brief reminders' restating sub-skill conventions, plus overlapping 'Problem-Solving Mode'/'Quick Lookup Mode'/'General Protocols' prose) and could be tightened.	2 / 3
Actionability	Provides copy-paste-ready commands ('tu run read_executed_notebook ...', 'Skill(skill="...")'), a tool table with concrete one-liners, and worked tie-break examples like 'calculate the drip rate for this IV -> computational-biophysics'.	3 / 3
Workflow Clarity	The routing process is clearly sequenced (FIRST ACTION, numbered How-to-route 1-4, explicit tie-break rules, and an 'If no signal matches' fallback) with correctness checkpoints in RULE ZERO (cite the cell output rather than recompute, prefer canonical over scratch-named scripts).	3 / 3
Progressive Disclosure	The body is an overview router pointing one level deep to ~70 sub-skills via Skill() calls, with a single clearly-signaled bundle reference (references/general-strategies.md) that exists on disk; details are appropriately externalized to the sub-skills.	3 / 3
	Total	11 / 12 Passed

Description

100%

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 concrete, trigger-rich, and answers both 'what' and 'when' with a distinct niche, scoring at the top of the rubric on all dimensions. Its main weakness is verbosity: the STEP 1 procedure embeds lengthy explanatory prose about outlier removal and library versions that pads the field well beyond what a concise description needs.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('run tu run read_executed_notebook ... to extract its cell outputs', 'pick a sub-skill name from this exact list') and enumerates specific sub-skill domains; not merely a domain label plus a few actions.	3 / 3
Completeness	Explicitly answers both what (router that reads executed notebooks and routes to sub-skills) and when ('Activate when the user provides data files to analyze...'), with an explicit 'Activate when' trigger clause.	3 / 3
Trigger Term Quality	Covers many natural trigger phrases a user would actually say via the when_to_use field ('asks about differential expression', 'does pathway enrichment', 'builds phylogenetic trees') plus concrete tool/domain keywords.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (ToolUniverse plugin router) with a fixed sub-skill list and specific file-signal triggers, making misfires for the wrong skill unlikely.	3 / 3
	Total	12 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: mims-harvard/ToolUniverse
Path: plugins/tooluniverse/skills/tooluniverse/SKILL.md
Commit: 089eb8e

Reviewed: about 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.