Drug Repurposing (Claude Skill)

A ToolUniverse agent skill that systematically identifies and prioritises drug-repurposing candidates by combining target-, compound-, and disease-driven reasoning with mechanism rationale, clinical-trial precedent, and regulatory feasibility.


Type	Claude Skill (one of ToolUniverse’s pre-built agent skills)
Supplier	Zitnik Lab, Harvard Medical School
Availability	GA — part of the ToolUniverse skills collection (`skills/tooluniverse-drug-repurposing/`)
Pricing	Free / OSS (Apache-2.0); wraps public APIs (Open Targets, ChEMBL, UniProt, Reactome, STRING)
Capabilities	Read-only — drives ToolUniverse tool calls; no data writes

How to install

This skill calls ToolUniverse tools, so the ToolUniverse MCP server must be installed first (see the ToolUniverse page). Simplest registration:

claude mcp add --transport stdio tooluniverse -- uvx tooluniverse

Then add the skills:

Claude Code — install the whole skill collection (the skill resolves as tooluniverse-drug-repurposing):
```
npx skills add mims-harvard/ToolUniverse
```

Manual / other agents — copy just this skill directory into your skills folder:

git clone https://github.com/mims-harvard/ToolUniverse
cp -r ToolUniverse/skills/tooluniverse-drug-repurposing ~/.claude/skills/

The skill sets disable-model-invocation: true upstream, so invoke it explicitly (e.g. ask Claude to “use the drug-repurposing skill”) rather than relying on automatic dispatch.

What it does

Guides a five-phase workflow — disease/target analysis, drug discovery, safety assessment, literature review, and composite scoring — across three strategies:

Target-based — identify disease-associated targets via genetic evidence (GWAS/rare variants), then find approved drugs that modulate them.
Compound-based — start from an approved drug, map its full primary/off-target profile via ChEMBL bioactivity, then search for new indications where secondary targets are relevant.
Disease-driven — start from a disease, derive associated targets/pathways via Reactome/STRING, then match to existing drugs.

Representative ToolUniverse tools called: OpenTargets_get_disease_id_description_by_name, OpenTargets_get_associated_targets_by_disease_efoId, UniProt_get_entry_by_accession, plus ChEMBL bioactivity and clinical-trial lookups.

Primary use cases: hypothesis-generating repurposing for orphan diseases, finding existing drugs for new indications, prioritising candidates by evidence and feasibility.

Notes

The skill emphasises mechanistic validation — each link (target involvement, drug potency, pathway overlap) must be verified with a tool call rather than assumed. It is a thin reasoning layer over ToolUniverse; without the MCP server registered, the tool calls fail. ToolUniverse ships ~68 such skills (drug research, target validation, drug-drug interaction, pharmacovigilance, precision oncology, rare-disease diagnosis); this entry covers the repurposing workflow specifically.

Sources

Installed this tool?

Share feedback — install path, OS, errors, workarounds. The form opens with this tool pre-selected and a link back to this page.