tooluniverse-network-pharmacology

Compound-target-disease network construction and analysis for drug repurposing, polypharmacology discovery, and multi-target drug design. Uses STRING, BioGRID, ChEMBL, DGIdb, OMIM, OpenTargets. Use for off-target effect prediction, network-based drug repurposing, and identifying molecules with desired multi-target profile.

Quality

75%

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 ./plugins/tooluniverse/skills/tooluniverse-network-pharmacology/SKILL.md

Quality

Content

50%

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

A well-structured, information-dense operational guide for an 8-phase network pharmacology pipeline with strong tool-parameter specifics, but it ships no in-body executable code and all six referenced detail files are missing from the bundle, undermining both actionability and progressive disclosure.

Suggestions

Include at least one concrete, executable Python example in the body (e.g., a Network_proximity invocation or a pandas Network Pharmacology Score computation) instead of deferring all code to ANALYSIS_PROCEDURES.md, and ensure the referenced .md files (ANALYSIS_PROCEDURES, REPORT_TEMPLATE, SCORING_REFERENCE, TOOL_REFERENCE, USE_PATTERNS, QUICK_START) actually ship in the bundle.

Convert the implicit principles into explicit validation checkpoints between phases, e.g. "Confirm every entity ID resolved non-empty before Phase 1" and "Verify C-T-D edge count > 0 before Phase 3 analysis", to add hard validate-then-proceed gates.

Tighten the verbose Network_proximity sub-section and consolidate the long per-phase tool lists / fallback chains into compact tables to cut tokens while preserving the operational detail.

Dimension	Reasoning	Score
Conciseness	The body avoids explaining concepts Claude already knows (no definition of PPI networks, STRING, or enrichment) and is information-dense, but it runs ~200 lines and sections like the multi-bullet Network_proximity block and the long per-phase tool lists could be tightened; not every token earns its place.	2 / 3
Actionability	Concrete guidance is strong — exact tool names, required params ("DrugBank tools: ALL require query, case_sensitive, exact_match, limit"), and thresholds ("Z < -2, p < 0.01", "IC50 < 1 uM") — but the body contains no executable Python code, deferring all examples to ANALYSIS_PROCEDURES.md, a file absent from the bundle, so copy-paste-ready code is missing.	2 / 3
Workflow Clarity	The 9-phase (Phase 0-8) sequence is explicit and the Fallback Strategies section gives retry chains, but validation is principle-based ("Entity disambiguation FIRST", "'No data' is data") rather than hard checkpoints; for a batch data-gathering pipeline the lack of explicit validate-before-proceeding gates caps this at 2.	2 / 3
Progressive Disclosure	The overview is well-organized with clearly signaled one-level-deep references ("> Full step-by-step code examples: ANALYSIS_PROCEDURES.md", plus a Reference Files section), but the six referenced .md files are not present in the bundle, so navigation is broken and the disclosure is not actually realized.	2 / 3
	Total	8 / 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.

A strong description that concisely states concrete capabilities, names the underlying data sources, and gives an explicit 'Use for' trigger clause covering the main use cases, with a clear specialized niche that minimizes conflict risk.

Dimension	Reasoning	Score
Specificity	Multiple concrete actions are named — "network construction and analysis", "polypharmacology discovery", "multi-target drug design", "off-target effect prediction" — matching the multiple-specific-actions anchor rather than the single-action score-2 example.	3 / 3
Completeness	It explicitly answers both what ("Compound-target-disease network construction and analysis for drug repurposing, polypharmacology discovery, and multi-target drug design") and when ("Use for off-target effect prediction, network-based drug repurposing..."), with an explicit trigger clause.	3 / 3
Trigger Term Quality	The "Use for" clause supplies natural domain terms a user would actually say — "off-target effect prediction", "network-based drug repurposing", "desired multi-target profile" — giving good coverage rather than jargon-only phrasing.	3 / 3
Distinctiveness Conflict Risk	The compound-target-disease / network-pharmacology niche with named databases (STRING, BioGRID, ChEMBL, DGIdb, OMIM, OpenTargets) is a clear, specialized niche with distinct triggers unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 10 missing, 5 suspicious	Warning

	Total	15 / 16 Passed

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

Reviewed: about 7 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.