Medical Research Toolkit

Query 14+ biomedical databases for drug repurposing, target discovery, clinical trials, and literature research. Access ChEMBL, PubMed, ClinicalTrials.gov, O...

Security Scan

The skill is an instruction-only documentation bundle that consistently describes and directs use of a third‑party unified biomedical API; it does not request extra credentials or install code, but it will send queries (and their contents) to an external endpoint — review that endpoint and avoid sending sensitive/PHI.

Purpose & Capability

Name/description match the content: SKILL.md and reference files consistently document queries against many biomedical databases via a single unified MCP endpoint. No unrelated credentials, binaries, or config paths are requested.

Instruction Scope

All runtime instructions are curl/json-rpc examples that POST queries to https://mcp.cloud.curiloo.com/tools/unified/mcp (or optionally to a local 'medical-mcps' service). The instructions do not ask the agent to read local files or hidden credentials, but they do direct potentially sensitive query text to an external third-party endpoint — this is expected for the stated purpose but has privacy implications (do not send patient-identifiable data).

Install Mechanism

No install spec or code is included in the skill bundle (instruction-only). The SKILL.md mentions 'pip install medical-mcps' as a local option, but that is an external package installation separate from the skill; the skill itself writes nothing to disk.

Credentials

The skill declares no required environment variables or credentials. Where APIs require keys (OMIM, optional OpenFDA/NCI), the docs show passing them as API parameters; the requested access is proportionate to the documented APIs. No unrelated secrets are requested.

Persistence & Privilege

The skill does not request always:true or elevated persistent privileges, and it does not modify agent/system configs. It is user-invocable and may be called autonomously by the agent (normal default).

Assessment

This skill is documentation and curl examples for a third‑party unified biomedical API; it appears internally consistent with that purpose. Before using it, verify the external endpoint (https://mcp.cloud.curiloo.com) and the upstream source (README links to a GitHub repo) so you trust who will receive your queries. Never send patient‑identifiable information or sensitive clinical data to the public endpoint. If you prefer local control, inspect the 'medical-mcps' package source on PyPI/GitHub before installing and consider running the service locally behind your network controls. Note small doc inconsistencies (OpenTargets uses EFO vs MONDO ids in different files) — validate returned IDs when chaining lookups. If you need a higher assurance review, ask for the upstream service code or a signed API/owner identity before sending sensitive queries.

Query 14+ biomedical databases for drug repurposing, target discovery, clinical evidence, and literature research — all via a unified MCP endpoint.

⚡ 30-Second Start

# Find drugs for myasthenia gravis
curl -X POST https://mcp.cloud.curiloo.com/tools/unified/mcp \
  -H "Content-Type: application/json" -H "Accept: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"chembl_find_drugs_by_indication","arguments":{"indication":"myasthenia gravis","max_results":10}},"id":1}'

That's it! You now have approved and investigational drugs for the disease.

---

Quick Recipes

Find Drugs for a Disease

curl -X POST https://mcp.cloud.curiloo.com/tools/unified/mcp \
  -H "Content-Type: application/json" -H "Accept: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"chembl_find_drugs_by_indication","arguments":{"indication":"myasthenia gravis","max_results":20}},"id":1}'

Returns: Approved + investigational drugs with max phase reached

Find Disease Targets

# First: Find disease ID
curl -X POST https://mcp.cloud.curiloo.com/tools/unified/mcp \
  -H "Content-Type: application/json" -H "Accept: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"opentargets_search","arguments":{"query":"myasthenia gravis","entity_type":"disease"}},"id":1}'

# Returns: disease ID (e.g., EFO_0004991)
# Then: Get targets
curl -X POST https://mcp.cloud.curiloo.com/tools/unified/mcp \
  -H "Content-Type: application/json" -H "Accept: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"opentargets_get_associations","arguments":{"disease_id":"EFO_0004991","size":20}},"id":2}'

Returns: Top disease targets ranked by evidence strength (0-1 score)

Search Literature

curl -X POST https://mcp.cloud.curiloo.com/tools/unified/mcp \
  -H "Content-Type: application/json" -H "Accept: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"pubmed_search_articles","arguments":{"diseases":["myasthenia gravis"],"keywords":["immunotherapy"],"max_results":20}},"id":1}'

Returns: PubMed articles on myasthenia gravis immunotherapy

Find Active Clinical Trials

curl -X POST https://mcp.cloud.curiloo.com/tools/unified/mcp \
  -H "Content-Type: application/json" -H "Accept: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"ctg_search_studies","arguments":{"condition":"myasthenia gravis","recruitment_status":"RECRUITING","max_results":20}},"id":1}'

Returns: Actively recruiting trials for the disease

Check Drug Safety

curl -X POST https://mcp.cloud.curiloo.com/tools/unified/mcp \
  -H "Content-Type: application/json" -H "Accept: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"openfda_search_adverse_events","arguments":{"drug_name":"eculizumab","limit":50}},"id":1}'

Returns: FDA adverse event reports—check for serious outcomes, death, hospitalization

---

What You Get

14+ Integrated Databases

Database	What's Inside	Use For
ChEMBL	2M drugs, bioactivity data	Finding approved/investigational drugs
OpenTargets	Disease-target links + evidence	Identifying therapeutic targets
PubMed	35M+ articles + preprints	Literature review & validation
ClinicalTrials.gov	400K+ active trials	Finding ongoing studies
OpenFDA	Adverse events, drug labels	Safety assessment
OMIM	Genetic diseases, genes → phenotypes	Understanding disease genetics
Reactome	Pathways, protein interactions	Mechanism understanding
UniProt	Protein sequences, annotations	Protein properties
KEGG	Metabolic & disease pathways	Systems-level view
GWAS Catalog	Genetic associations	Variant discovery
Pathway Commons	Integrated pathway data	Network analysis
MyGene.info	Gene annotations	ID mapping
MyVariant.info	Variant effects	Variant interpretation
+ more

---

Use Cases

🧬 Drug Repurposing

Find non-standard-care drugs for rare/complex diseases:

1. Find disease targets (OpenTargets)
2. Search for drugs targeting those genes (ChEMBL)
3. Check ongoing trials (ClinicalTrials)
4. Verify safety (OpenFDA, PubMed)

🔬 Target Discovery

Identify novel therapeutic targets:

1. Find disease associations (OpenTargets, GWAS)
2. Get pathway context (Reactome, KEGG)
3. Review literature (PubMed)
4. Check protein properties (UniProt)

📋 Clinical Evidence Review

Compile evidence for a hypothesis:

1. Search trials (ClinicalTrials.gov)
2. Find literature (PubMed)
3. Check FDA data (OpenFDA)

📊 Literature Mining

Systematically search biomedical research:

1. PubMed: 35M+ articles searchable by gene, disease, drug, chemical
2. Preprints: bioRxiv, medRxiv
3. Filter by keywords, date, study type

---

API Endpoint

Production (No setup needed):

https://mcp.cloud.curiloo.com/tools/unified/mcp

All 14+ databases unified into one endpoint.

Running Locally:

pip install medical-mcps
medical-mcps
# Available at: http://localhost:8000/tools/unified/mcp

---

Complete References

See detailed guides for each database:

• PubMed Guide — Literature search (genes, diseases, keywords)
• ClinicalTrials Guide — Find active trials
• ChEMBL Guide — Drug-target data & bioactivity
• OpenTargets Guide — Disease-target associations
• OpenFDA Guide — Drug safety & adverse events
• OMIM Guide — Genetic diseases (requires API key)
• Other APIs — Reactome, UniProt, KEGG, GWAS, etc.

---

Workflow Example

Complete Drug Repurposing Pipeline:

See drug-repurposing-workflow.md for step-by-step example with all 8 steps + curl commands.

---

API Keys

Most APIs are free, no key required. Optional keys for higher rate limits:

Database	Key?	Why	Get Key
ChEMBL	No	Public data	(not needed)
OpenTargets	No	Public data	(not needed)
PubMed	No	Public data	(not needed)
ClinicalTrials	No	Public data	(not needed)
OMIM	Yes	Proprietary data	https://omim.org/api
OpenFDA	Optional	Higher rate limits	https://open.fda.gov
NCI Clinical Trials	Optional	Higher rate limits	https://clinicaltrialsapi.cancer.gov

---

Rate Limits & Caching

• No authentication for production endpoint (public)
• Rate limits: Generous (~1000+ requests/day per database)
• Caching: Automatic 30-day HTTP caching (RFC 9111)
• Cost: $0 (all databases public or researcher-accessible)

---

Common Patterns

Batch Query Loop

# Search multiple targets
for gene in CHRNE RAPSN LRP4; do
  curl -X POST https://mcp.cloud.curiloo.com/tools/unified/mcp \
    -H "Content-Type: application/json" -H "Accept: application/json" \
    -d "{\"jsonrpc\":\"2.0\",\"method\":\"tools/call\",\"params\":{\"name\":\"chembl_find_drugs_by_target\",\"arguments\":{\"target_name\":\"$gene\",\"max_results\":10}},\"id\":1}"
  sleep 1  # Be nice to the API
done

ID Conversion

Need to convert IDs between databases?

curl -X POST https://mcp.cloud.curiloo.com/tools/unified/mcp \
  -H "Content-Type: application/json" -H "Accept: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"nodenorm_get_normalized_nodes","arguments":{"curie":"HGNC:11998"}},"id":1}'

Returns: Equivalent IDs in NCBI Gene, Uniprot, Ensembl, MyGene, etc.

---

Troubleshooting

No results?

• Try alternative terms (gene symbol vs protein name)
• Check spelling
• Use standardized IDs (gene symbols, disease names)
• Some APIs use specific ID formats (EFO vs MONDO, etc.)

Too many results?

• Add filters (max_results, phase, recruitment status)
• Use limit or size parameters
• Combine with other databases to narrow focus

API key errors?

• OMIM requires API key — get from https://omim.org/api
• Other databases optional — request key if hitting rate limits

---

Next Steps

1. Pick a use case (drug repurposing, target discovery, etc.)
2. Read the relevant database guide from References section
3. Copy a quick recipe from above
4. Customize parameters for your disease/gene/drug
5. Scale up — build your full research workflow

---

Resources

• Source Code: https://github.com/pascalwhoop/medical-mcps
• Author: Pascal Brockmeyer (@pascalwhoop)
• Organization: Every Cure (https://www.everycure.org)
• License: MIT

---

Getting Help

• Database not working? → Check Troubleshooting
• Want detailed guide? → See Complete References
• Need a workflow example? → See drug-repurposing-workflow.md
• Question about OpenClaw? → See OPENCLAW-USAGE.md