Skip to main content
Glama

BioMCP

One binary. One grammar. Evidence from the biomedical sources you already trust.

Data terms

BioMCP retrieves data from upstream biomedical sources. Those sources' terms govern how you use the data they return: some restrict commercial or clinical use. Check each source's terms and obtain any licence you need before relying on the data. See Source Licensing and Terms for the per-source breakdown.

Related MCP server: clinicaltrialsgov-mcp-server

What is BioMCP?

BioMCP is one CLI binary over a single command grammar that reaches 72 trusted biomedical sources directly (PubMed, ClinVar, ClinicalTrials.gov, OncoKB, Reactome, and more; ten further sources arrive inside another source's answer). It is also an MCP (Model Context Protocol) server, so the same tools are available to AI agents such as Claude Code, Codex, and Claude Desktop.

BioMCP cuts through the usual biomedical data maze: one query reaches the sources that normally live behind different APIs, identifiers, and search habits. Researchers, clinicians, and agents use the same command grammar to search, focus, and pivot without rebuilding the workflow for each source. You get compact, evidence-oriented results across live public data plus local study analytics.

Features

  • Search the literature: search article fans out across PubTator3 and Europe PMC and PubMed when the filters are compatible, deduplicates PMID/PMCID/DOI identifiers, and can add a Semantic Scholar leg when your filters support it.

  • Pivot without rework: move from a gene, variant, drug, disease, pathway, protein, or article straight into the next built-in view instead of rebuilding filters by hand.

  • Choose a playbook: biomcp skill list shows shipped worked examples so you can open the matching biomcp skill <slug> workflow.

  • Analyze studies locally: study commands cover local query, cohort, survival, compare, and co-occurrence workflows with native terminal, SVG, and PNG charts for downloaded cBioPortal-style datasets.

  • Follow the paper trail: article citations, article references, article recommendations, and article entities turn one known paper into a broader evidence map.

  • Enrich and batch: use biomcp enrich for top-level g:Profiler enrichment and biomcp batch for up to 10 focused get calls in one command.

Quick start

First useful query in under 30 seconds:

uv tool install biomcp-cli
biomcp health --apis-only
biomcp skill list
biomcp list gene
biomcp search all --gene BRAF --disease melanoma  # unified cross-entity discovery
biomcp get gene BRAF pathways hpa

Watch the introduction

A project using BioMCP won St. Jude's KIDS BioHackathon in 2025. The talk that followed is a one-hour introduction to BioMCP, recorded October 24, 2025.

BioMCP: An Introduction to Biomedical AI Agents

Watch on YouTube &middot; Read the write-up &middot; Full transcript

Installation

Binary install

curl -fsSL https://biomcp.org/install.sh | bash

PyPI tool install

uv tool install biomcp-cli
# or: pip install biomcp-cli

PyPI package warning: install biomcp-cli, not biomcp. The biomcp PyPI package is unrelated to this project.

MCP Registry ownership marker: mcp-name: io.github.genomoncology/biomcp.

This installs the biomcp binary in ~/.local/bin. If that directory is not already on PATH, the installer prints one command to add it; it never edits your shell startup files.

Homebrew

brew tap genomoncology/biomcp
brew install biomcp

The separate genomoncology/homebrew-biomcp tap repository must exist before these commands can work.

Docker

docker run --rm ghcr.io/genomoncology/biomcp --version
docker run --rm ghcr.io/genomoncology/biomcp list
docker run --rm -i ghcr.io/genomoncology/biomcp serve

Use the GHCR image for quick CLI checks or stdio MCP clients without a local install.

Claude Code plugin

Install the biomcp binary first, then add the hosted plugin marketplace and install the BioMCP plugin in Claude Code:

/plugin marketplace add genomoncology/biomcp
/plugin install biomcp@biomcp

The plugin wires Claude Code to the local stdio MCP server with biomcp serve. For guided BioMCP workflows, also install the skill assets below.

Codex MCP server

Install the biomcp binary first, then register the same stdio MCP server with Codex:

codex mcp add biomcp -- biomcp serve

Claude Desktop extension (.mcpb)

Install BioMCP from the Anthropic Directory in Claude Desktop when that path is available for your environment. For local/manual setups, use the JSON MCP config below.

Install skills

The executable embeds the guide and worked examples. Read the complete guide with biomcp skill render, or install copies into your agent directory:

biomcp skill install ~/.claude --force

MCP clients

{
  "mcpServers": {
    "biomcp": {
      "command": "biomcp",
      "args": ["serve"]
    }
  }
}

Remote HTTP server

For shared or remote deployments:

biomcp serve-http --host 127.0.0.1 --port 8080

Remote clients connect to http://127.0.0.1:8080/mcp. Probe routes are GET /health, GET /readyz, and GET /.

Runnable demo:

uv run --script examples/streamable-http/streamable_http_client.py

See Remote HTTP Server for the newcomer guide.

From source

make install
"$HOME/.local/bin/biomcp" --version

For repo-local verification, run the standard gates directly: make lint, make test, and make spec. make test includes both Rust nextest and the Python/docs contract lane, while make release-gate adds the named full-feature proof and runs specs against the all-feature release binary. There is no supported make check command. Use make verify only for opt-in live public-upstream confidence; make release-live-smoke remains a compatibility alias.

Command grammar

search <entity> [filters]    → discovery
skill list                   → playbook catalog for how-to questions
discover <query>             → concept resolution before entity selection
get <entity> <id> [sections] → focused detail
<entity> <helper> <id>       → cross-entity pivots
enrich <GENE1,GENE2,...>     → gene-set enrichment
batch <entity> <id1,id2,...> → parallel gets
search all [slot filters]    → counts-first cross-entity orientation

Entities and sources

The tables below distinguish detail-card entities from search-only surfaces so agents do not synthesize unsupported get commands.

Gettable entities

Entity

Upstream providers used by BioMCP

Example

gene

MyGene.info, UniProt, Reactome, QuickGO, STRING, GTEx, Human Protein Atlas, DGIdb, ClinGen, NIH Reporter, DisGeNET, GTR-backed diagnostics pivot

biomcp get gene BRAF pathways hpa

variant

MyVariant.info, ClinVar, direct gnomAD v4 population data, CIViC, Cancer Genome Interpreter, OncoKB, cBioPortal, GWAS Catalog, AlphaGenome

biomcp get variant "BRAF V600E" clinvar

article

PubMed, PubTator3, Europe PMC, PMC OA, NCBI ID Converter, Semantic Scholar (optional auth; S2_API_KEY recommended)

biomcp search article -g BRAF --limit 5

author

Semantic Scholar, ORCID

biomcp get author semanticscholar:1716151

trial

ClinicalTrials.gov API v2, NCI CTS API

biomcp search trial -c melanoma -s recruiting

diagnostic

NCBI Genetic Testing Registry local bulk bundle + WHO IVD local CSV + optional OpenFDA device overlay

biomcp get diagnostic GTR000006692.3 regulatory

drug

MyChem.info, DDInter local bundle, EMA local batch, WHO Prequalification local exports, ChEMBL, OpenTargets, Drugs@FDA, OpenFDA labels/shortages/approvals/FAERS/MAUDE/recalls, CIViC

biomcp drug interactions warfarin

disease

MyDisease.info, Monarch Initiative, MONDO, OpenTargets, Reactome, CIViC, SEER Explorer, NIH Reporter, DisGeNET, GTR/WHO IVD diagnostics pivot

biomcp get disease "Lynch syndrome" genes

pathway

Reactome, KEGG, WikiPathways, g:Profiler, Enrichr-backed enrichment sections

biomcp get pathway hsa05200 genes

protein

UniProt, InterPro, STRING, ComplexPortal, PDB, AlphaFold

biomcp get protein P15056 complexes

adverse-event

OpenFDA FAERS/MAUDE/recalls plus CDC WONDER VAERS aggregate vaccine search

biomcp search adverse-event --drug pembrolizumab

pgx

CPIC, PharmGKB

biomcp get pgx CYP2D6 recommendations

cell-line

Cellosaurus with ChEMBL, PharmacoDB, DepMap, COSMIC, GDSC, and LINCS cross-references

biomcp get cell-line CVCL_0004 xrefs

Search-only entities

Entity

Upstream providers used by BioMCP

Example

gwas

GWAS Catalog

biomcp search gwas --trait "type 2 diabetes"

phenotype

Monarch Initiative (HPO semantic similarity plus direct-support checks)

biomcp search phenotype "HP:0001250"

Cross-entity helpers

Pivot between related entities without rebuilding filters.

See the cross-entity pivot guide for when to use a helper versus a fresh search.

biomcp variant trials "BRAF V600E" --limit 5
biomcp variant articles "BRAF V600E"
biomcp drug adverse-events pembrolizumab
biomcp drug trials pembrolizumab
biomcp disease trials melanoma
biomcp disease drugs melanoma
biomcp disease articles "Lynch syndrome"
biomcp gene trials BRAF
biomcp gene drugs BRAF
biomcp gene articles BRCA1
biomcp gene pathways BRAF
biomcp pathway drugs R-HSA-5673001
biomcp pathway drugs hsa05200
biomcp pathway articles R-HSA-5673001
biomcp pathway trials R-HSA-5673001
biomcp protein structures P15056
biomcp article entities 22663011
biomcp article citations 22663011 --limit 3
biomcp article references 22663011 --limit 3
biomcp article recommendations 22663011 --limit 3

Gene-set enrichment

biomcp enrich BRAF,KRAS,NRAS --limit 10

Top-level biomcp enrich uses g:Profiler. Gene enrichment sections inside other entity views still reference Enrichr where that is the backing source.

Sections and progressive disclosure

Every get command supports selectable sections for focused output:

biomcp get gene BRAF                    # summary card
biomcp get gene BRAF pathways           # add pathway section
biomcp get gene BRCA1 diagnostics       # diagnostic-test pivot from GTR
biomcp get gene BRAF hpa                # protein tissue expression + localization
biomcp get gene BRAF civic interactions # multiple sections
biomcp get gene BRAF all                # standard sections; diagnostics/funding stay opt-in

biomcp get variant "BRAF V600E" clinvar population conservation
biomcp get article 22663011 tldr
biomcp get drug pembrolizumab label targets civic approvals
biomcp get drug trastuzumab regulatory --region who
biomcp get disease "Lynch syndrome" genes phenotypes variants
biomcp get disease tuberculosis diagnostics
biomcp get diagnostic GTR000006692.3 regulatory
biomcp get trial NCT02576665 eligibility locations outcomes

In JSON mode, get responses expose _meta.next_commands for the next likely follow-ups and _meta.section_sources for section-level provenance. batch ... --json returns per-entity objects with the same metadata shape.

API keys

Most commands work without credentials. Optional keys improve rate limits or unlock optional enrichments:

export NCBI_API_KEY="..."        # ClinVar, PubTator, PubMed/efetch, PMC OA, NCBI ID converter
export S2_API_KEY="..."          # Optional Semantic Scholar auth; dedicated quota at 1 req/sec
export OPENFDA_API_KEY="..."     # OpenFDA rate limits
export NCI_API_KEY="..."         # NCI CTS trial search (--source nci)
export ONCOKB_TOKEN="..."        # OncoKB variant helper
export ALPHAGENOME_API_KEY="..." # AlphaGenome variant effect prediction

search article, get article, batch article, get article ... tldr, and the explicit Semantic Scholar helpers all work without S2_API_KEY. With the key, BioMCP sends authenticated requests and uses a dedicated rate limit at 1 req/sec. Without it, BioMCP uses the shared unauthenticated pool at 1 req/2sec. search article --source supports all, pubtator, europepmc, pubmed, semanticscholar, and litsense2. The default compatible article federation uses PubTator3, Europe PMC, PubMed, and automatic Semantic Scholar; use --source semanticscholar or --source litsense2 explicitly when you want one of those sources alone. Explicit source selection also disables cross-provider row enrichment. References and recommendations can be empty for paywalled papers because of publisher elision in Semantic Scholar upstream coverage.

Configuration

Claude Desktop extension settings

The directory bundle exposes only the optional settings needed for the first reviewer-facing build:

Claude Desktop field

Runtime env var

Purpose

OncoKB Token

ONCOKB_TOKEN

Enables biomcp variant oncokb "<gene> <variant>" therapy and level evidence

DisGeNET API Key

DISGENET_API_KEY

Enables scored DisGeNET sections on gene and disease lookups

Semantic Scholar API Key

S2_API_KEY

Improves reliability for article TLDR, citation, reference, and recommendation helpers

The first directory build exposes only those three optional settings. Advanced CLI-only env vars remain documented in API Keys for the general BioMCP CLI path.

Usage Examples

Public cross-entity overview

User prompt: Give me a low-noise overview of BRAF in melanoma.

Expected tool call: biomcp search all --gene BRAF --disease melanoma --counts-only

Expected behavior: Returns a cross-entity counts summary that orients the next command instead of dumping long detail tables.

Expected output: Counts-first summary with suggested next commands for the highest-yield entity follow-ups.

Public variant evidence

User prompt: Summarize ClinVar significance and population frequency for BRAF V600E.

Expected tool call: biomcp get variant "BRAF V600E" clinvar population

Expected behavior: Retrieves the focused variant card, ClinVar section, and population-frequency data in one read-only call.

Expected output: Variant summary, ClinVar significance details, and gnomAD population frequencies.

Credentialed OncoKB example

User prompt: Show OncoKB therapy evidence for BRAF V600E.

Expected tool call: biomcp variant oncokb "BRAF V600E"

Expected behavior: Uses ONCOKB_TOKEN when configured and otherwise returns helpful guidance about the missing credential.

Expected output: Therapy and level evidence when ONCOKB_TOKEN is set, or a clear setup hint when it is not.

Credentialed DisGeNET example

User prompt: Show scored DisGeNET associations for TP53.

Expected tool call: biomcp get gene TP53 disgenet

Expected behavior: Uses DISGENET_API_KEY to retrieve the scored gene-disease association section.

Expected output: Ranked disease-association table with evidence counts and scores when DISGENET_API_KEY is configured.

Privacy Policy

BioMCP does not add telemetry, analytics, or remote log upload. Review the full privacy statement at https://biomcp.org/policies/.

Multi-worker deployment

BioMCP rate limiting is process-local. For many concurrent workers, run one shared Streamable HTTP biomcp serve-http endpoint so all workers share a single limiter budget:

biomcp serve-http --host 0.0.0.0 --port 8080 \
  --allowed-hosts biomcp.example.org

Loopback servers accept only local Host values by default. A non-loopback bind requires --allowed-hosts. The explicit --unsafe-allow-any-host escape hatch disables only this Host check; it does not add authentication, TLS, or encryption. Put remote deployments behind a trusted authenticated TLS proxy or inside a private network.

Remote clients should connect to http://<host>:8080/mcp. Lightweight process probes are available at GET /health, GET /readyz, and GET /.

Skills

BioMCP ships an embedded agent guide and a worked-example catalog. Use biomcp skill list when you need the right worked example, then use biomcp skill to read the embedded BioMCP guide or install it into your agent directory when you want local copies of the workflow references:

biomcp skill list
biomcp skill
biomcp skill install ~/.claude --force

See Skills for supported install targets, installed files, and legacy compatibility notes.

Local study analytics

study is BioMCP's local analysis family for downloaded cBioPortal-style datasets. The public entity surface handles API-backed, local-runtime, and hybrid discovery/detail; study commands work on local datasets when you need per-study query, cohort, survival, comparison, or co-occurrence workflows. Per-gene queries include mutations, CNA, expression, and structural variants/fusions from local data_sv.txt files. Mutation summaries stay mutation-only and note when fusions/SV need --type sv.

Use study download to fetch a dataset into your local study root. Set BIOMCP_STUDY_DIR when you want an explicit dataset location for reproducible scripts and demos; if it is unset, BioMCP falls back to its default study root.

export BIOMCP_STUDY_DIR="$HOME/.local/share/biomcp/studies"
biomcp study download msk_impact_2017
biomcp study query --study msk_impact_2017 --gene TP53 --type mutations --chart bar --theme dark --palette wong -o docs/blog/images/tp53-mutation-bar.svg
biomcp study query --study msk_impact_2017 --gene RET --type sv

See the CLI reference for the full study command family and dataset prerequisites.

Ops

biomcp version                            # show version and build info
biomcp health                             # inspect API connectivity plus local DDInter/EMA/cache readiness
biomcp update                             # self-update with release SHA256 checksum verification
biomcp update --check                     # check for updates without installing
biomcp uninstall                          # remove biomcp from ~/.local/bin

Support

Documentation

Citation

If you use BioMCP in research, cite it via CITATION.cff. GitHub also exposes Cite this repository in the repository sidebar when that file is present.

Data Sources and Licensing

BioMCP is MIT-licensed. It performs on-demand queries against upstream providers instead of vendoring or mirroring their datasets, but upstream terms govern reuse of retrieved results.

Some providers are fully open, some BioMCP features require registration or API keys, and some queryable sources still impose notable reuse limits. The two biggest cautions are KEGG, which distinguishes academic and non-academic use, and COSMIC, which BioMCP keeps indirect-only because its licensing model is incompatible with a direct open integration.

Use Source Licensing and Terms for the per-source breakdown and API Keys for setup steps and registration links.

License

MIT

Available Tools

35 tools
alphagenome_predictorA

Predict variant effects on gene regulation using Google DeepMind's AlphaGenome.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your analysis strategy!

AlphaGenome provides state-of-the-art predictions for how genetic variants
affect gene regulation, including:
- Gene expression changes (RNA-seq)
- Chromatin accessibility impacts (ATAC-seq, DNase-seq)
- Splicing alterations
- Promoter activity changes (CAGE)

This tool requires:
1. AlphaGenome to be installed (see error message for instructions)
2. An API key from https://deepmind.google.com/science/alphagenome

API Key Options:
- Provide directly via the api_key parameter
- Or set ALPHAGENOME_API_KEY environment variable

Example usage:
- Predict regulatory effects of BRAF V600E mutation: chr7:140753336 A>T
- Assess non-coding variant impact on gene expression
- Evaluate promoter variants in specific tissues

Note: This is an optional tool that enhances variant interpretation
with AI predictions. Standard annotations remain available via variant_getter.

Inputs: alternate (required), chromosome (required), position (required), reference (required), +4 more. Returns: result

article_getterA

Fetch detailed information for a specific article.

Retrieves the full abstract and available text for an article by its identifier.
Supports:
- PubMed IDs (PMID) for published articles
- PMC IDs for articles in PubMed Central
- DOIs for preprints from Europe PMC

Returns formatted text including:
- Title
- Abstract
- Full text (when available from PMC for published articles)
- Source information (PubMed or Europe PMC)

Inputs: pmid (required). Returns: result

article_searcherA

Search PubMed/PubTator3 for research articles and preprints.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your research strategy!

Use this tool to find scientific literature ABOUT genes, variants, diseases, or chemicals.
Results include articles from PubMed and optionally preprints from bioRxiv/medRxiv.

Important: This searches for ARTICLES ABOUT these topics, not database records.
For genetic variant database records, use variant_searcher instead.

Example usage:
- Find articles about BRAF mutations in melanoma
- Search for papers on a specific drug's effects
- Locate research on gene-disease associations

Inputs: chemicals, diseases, genes, include_cbioportal, +5 more. Returns: result

disease_getterA

Get detailed disease information from MyDisease.info.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to understand your research goal!

Provides real-time disease annotations including:
- Official disease name and definition
- Disease synonyms and alternative names
- Ontology mappings (MONDO, DOID, OMIM, etc.)
- Associated phenotypes
- Links to disease databases

This tool fetches CURRENT disease information from MyDisease.info, ensuring
you always have the latest ontology mappings and definitions.

Example usage:
- Get the definition of GIST (Gastrointestinal Stromal Tumor)
- Look up synonyms for melanoma
- Find the MONDO ID for a disease by name

Note: For clinical trials about diseases, use trial_searcher. For articles about diseases, use article_searcher.

Inputs: disease_id_or_name (required). Returns: result

drug_getterA

Get detailed drug/chemical information from MyChem.info.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to understand your research goal!

This tool provides comprehensive drug information including:
- Chemical properties (formula, InChIKey)
- Drug identifiers (DrugBank, ChEMBL, PubChem)
- Trade names and brand names
- Clinical indications
- Mechanism of action
- Pharmacology details
- Links to drug databases

This tool fetches CURRENT drug information from MyChem.info, part of the
BioThings suite, ensuring you always have the latest drug data.

Example usage:
- Get information about imatinib (Gleevec)
- Look up details for DrugBank ID DB00619
- Find the mechanism of action for pembrolizumab

Note: For clinical trials about drugs, use trial_searcher. For articles about drugs, use article_searcher.

Inputs: drug_id_or_name (required). Returns: result

fetchA

Fetch comprehensive details for a specific biomedical record.

This tool retrieves full information for articles, clinical trials, genetic variants,
genes, drugs, or diseases using their unique identifiers. It returns data in a
standardized format suitable for detailed analysis and research.

## IDENTIFIER FORMATS:
- Articles: PMID (PubMed ID) - e.g., "35271234" OR DOI - e.g., "10.1101/2024.01.20.23288905"
- Trials: NCT ID (ClinicalTrials.gov ID) - e.g., "NCT04280705"
- Variants: HGVS notation or dbSNP ID - e.g., "chr7:g.140453136A>T" or "rs121913254"
- Genes: Gene symbol or Entrez ID - e.g., "BRAF" or "673"
- Drugs: Drug name or ID - e.g., "imatinib" or "DB00619"
- Diseases: Disease name or ID - e.g., "melanoma" or "MONDO:0005105"
- NCI Organizations: NCI organization ID - e.g., "NCI-2011-03337"
- NCI Interventions: NCI intervention ID - e.g., "INT123456"
- NCI Diseases: NCI disease ID - e.g., "C4872"

The domain is automatically detected from the ID format if not provided:
- NCT* → trial
- Contains "/" with numeric prefix (DOI) → article
- Pure numeric → article (PMID)
- rs* or contains ':' or 'g.' → variant
- For genes, drugs, diseases: manual specification recommended

## DOMAIN-SPECIFIC OPTIONS:

### Articles (domain="article"):
- Returns full article metadata, abstract, and full text when available
- Supports both PubMed articles (via PMID) and Europe PMC preprints (via DOI)
- Includes annotations for genes, diseases, chemicals, and variants (PubMed only)
- detail="full" attempts to retrieve full text content (PubMed only)

### Clinical Trials (domain="trial"):
- detail=None or "protocol": Core study information
- detail="locations": Study sites and contact information
- detail="outcomes": Primary/secondary outcomes and results
- detail="references": Related publications and citations
- detail="all": Complete trial record with all sections

### Variants (domain="variant"):
- Returns comprehensive variant information including:
  - Clinical significance and interpretations
  - Population frequencies
  - Gene/protein effects
  - External database links
- detail parameter is ignored (always returns full data)

### Genes (domain="gene"):
- Returns gene information from MyGene.info including:
  - Gene symbol, name, and type
  - Entrez ID and Ensembl IDs
  - Gene summary and aliases
  - RefSeq information
- detail parameter is ignored (always returns full data)

### Drugs (domain="drug"):
- Returns drug/chemical information from MyChem.info including:
  - Drug name and trade names
  - Chemical formula and structure IDs
  - Clinical indications
  - Mechanism of action
  - External database links (DrugBank, PubChem, ChEMBL)
- detail parameter is ignored (always returns full data)

### Diseases (domain="disease"):
- Returns disease information from MyDisease.info including:
  - Disease name and definition
  - MONDO ontology ID
  - Disease synonyms
  - Cross-references to other databases
  - Associated phenotypes
- detail parameter is ignored (always returns full data)

### NCI Organizations (domain="nci_organization"):
- Returns organization information from NCI database including:
  - Organization name and type
  - Full address and contact information
  - Research focus areas
  - Associated clinical trials
- Requires NCI API key
- detail parameter is ignored (always returns full data)

### NCI Interventions (domain="nci_intervention"):
- Returns intervention information from NCI database including:
  - Intervention name and type
  - Synonyms and alternative names
  - Mechanism of action (for drugs)
  - FDA approval status
  - Associated clinical trials
- Requires NCI API key
- detail parameter is ignored (always returns full data)

### NCI Diseases (domain="nci_disease"):
- Returns disease information from NCI controlled vocabulary including:
  - Preferred disease name
  - Disease category and classification
  - All known synonyms
  - Cross-reference codes (ICD, SNOMED)
- Requires NCI API key
- detail parameter is ignored (always returns full data)

## RETURN FORMAT:
All fetch operations return a standardized format:
```json
{
    "id": "unique_identifier",
    "title": "Record title or name",
    "text": "Full content or comprehensive description",
    "url": "Link to original source",
    "metadata": {
        // Domain-specific additional fields
    }
}
```

## EXAMPLES:

Fetch article by PMID (domain auto-detected):
```
await fetch(id="35271234")
```

Fetch article by DOI (domain auto-detected):
```
await fetch(id="10.1101/2024.01.20.23288905")
```

Fetch complete trial information (domain auto-detected):
```
await fetch(
    id="NCT04280705",
    detail="all"
)
```

Fetch variant with clinical interpretations:
```
await fetch(id="rs121913254")
```

Explicitly specify domain (optional):
```
await fetch(
    domain="variant",
    id="chr7:g.140453136A>T"
)
```

Inputs: id (required), api_key, call_benefit, detail, +1 more.

gene_getterA

Get detailed gene information from MyGene.info.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to understand your research goal!

Provides real-time gene annotations including:
- Official gene name and symbol
- Gene summary/description
- Aliases and alternative names
- Gene type (protein-coding, etc.)
- Links to external databases

This tool fetches CURRENT gene information from MyGene.info, ensuring
you always have the latest annotations and nomenclature.

Example usage:
- Get information about TP53 tumor suppressor
- Look up BRAF kinase gene details
- Find the official name for a gene by its alias

Note: For genetic variants, use variant_searcher. For articles about genes, use article_searcher.

Inputs: gene_id_or_symbol (required). Returns: result

nci_biomarker_searcherA

Search for biomarkers in the NCI Clinical Trials database.

Searches for biomarkers used in clinical trial eligibility criteria.
This is essential for precision medicine trials that select patients
based on specific biomarker characteristics.

Biomarker examples:
- Gene mutations (e.g., BRAF V600E, EGFR T790M)
- Protein expression (e.g., PD-L1 ≥ 50%, HER2 positive)
- Gene fusions (e.g., ALK fusion, ROS1 fusion)
- Other molecular markers (e.g., MSI-H, TMB-high)

Requires NCI API key from: https://clinicaltrialsapi.cancer.gov/

Note: Biomarker data availability may be limited in CTRP.
Results focus on biomarkers used in trial eligibility criteria.

Example usage:
- Search for PD-L1 expression biomarkers
- Find trials requiring EGFR mutations
- Look up biomarkers tested by NGS
- Search for HER2 expression markers

Inputs: api_key, biomarker_type, name, page, +1 more. Returns: result

nci_disease_searcherA

Search NCI's controlled vocabulary of cancer conditions.

Searches the National Cancer Institute's curated database of cancer
conditions and diseases used in clinical trials. This is different from
the general disease_getter tool which uses MyDisease.info.

NCI's disease vocabulary provides:
- Official cancer terminology used in trials
- Disease synonyms and alternative names
- Hierarchical disease classifications
- Standardized disease codes for trial matching

Requires NCI API key from: https://clinicaltrialsapi.cancer.gov/

Example usage:
- Search for specific cancer types (e.g., "melanoma")
- Find all lung cancer subtypes
- Look up official names for disease synonyms
- Get standardized disease terms for trial searches

Note: This is specifically for NCI's cancer disease vocabulary.
For general disease information, use the disease_getter tool.

Inputs: api_key, category, include_synonyms, name, +2 more. Returns: result

nci_intervention_getterA

Get detailed information about a specific intervention from NCI.

Retrieves comprehensive details about an intervention including:
- Full name and synonyms
- Intervention type and category
- Mechanism of action (for drugs)
- FDA approval status
- Associated clinical trials
- Combination therapies

Requires NCI API key from: https://clinicaltrialsapi.cancer.gov/

Example usage:
- Get details about a specific drug
- Find all trials using a device
- View combination therapy protocols

Inputs: intervention_id (required), api_key. Returns: result

nci_intervention_searcherA

Search for interventions in the NCI Clinical Trials database.

Searches the National Cancer Institute's curated database of interventions
used in cancer clinical trials. This includes:
- FDA-approved drugs
- Investigational agents
- Medical devices
- Surgical procedures
- Radiation therapies
- Behavioral interventions

Requires NCI API key from: https://clinicaltrialsapi.cancer.gov/

Example usage:
- Find all trials using pembrolizumab
- Search for CAR-T cell therapies
- List radiation therapy protocols
- Find dietary interventions

Inputs: api_key, intervention_type, name, page, +2 more. Returns: result

nci_organization_getterA

Get detailed information about a specific organization from NCI.

Retrieves comprehensive details about an organization including:
- Full name and aliases
- Address and contact information
- Organization type and role
- Associated clinical trials
- Research focus areas

Requires NCI API key from: https://clinicaltrialsapi.cancer.gov/

Example usage:
- Get details about a specific cancer center
- Find contact information for trial sponsors
- View organization's trial portfolio

Inputs: organization_id (required), api_key. Returns: result

nci_organization_searcherA

Search for organizations in the NCI Clinical Trials database.

Searches the National Cancer Institute's curated database of organizations
involved in cancer clinical trials. This includes:
- Academic medical centers
- Community hospitals
- Industry sponsors
- Government facilities
- Research networks

Requires NCI API key from: https://clinicaltrialsapi.cancer.gov/

IMPORTANT: To avoid API errors, always use city AND state together when searching by location.
The NCI API has limitations on broad searches.

Example usage:
- Find cancer centers in Boston, MA (city AND state)
- Search for "MD Anderson" in Houston, TX
- List academic organizations in Cleveland, OH
- Search by organization name alone (without location)

Inputs: api_key, city, name, organization_type, +3 more. Returns: result

openfda_adverse_getterA

Get detailed information for a specific FDA adverse event report.

Retrieves complete details including:
- Patient demographics and medical history
- All drugs involved and dosages
- Complete list of adverse reactions
- Event narrative and outcomes
- Reporter information

Inputs: report_id (required), api_key. Returns: result

openfda_adverse_searcherA

Search FDA adverse event reports (FAERS) for drug safety information.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your research strategy!

Searches FDA's Adverse Event Reporting System for:
- Drug side effects and adverse reactions
- Serious event reports (death, hospitalization, disability)
- Safety signal patterns across patient populations

Note: These reports do not establish causation - they are voluntary reports
that may contain incomplete or unverified information.

Inputs: api_key, drug, limit, page, +2 more. Returns: result

openfda_approval_getterA

Get detailed FDA drug approval information for a specific application.

Returns comprehensive approval details including:
- Full product list with dosage forms and strengths
- Complete submission history
- Marketing status timeline
- Therapeutic equivalence codes
- Pharmacologic class information

Inputs: application_number (required), api_key. Returns: result

openfda_approval_searcherA

Search FDA drug approval records from Drugs@FDA database.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your research strategy!

Returns information about:
- Application numbers and sponsors
- Brand and generic names
- Product formulations and strengths
- Marketing status and approval dates
- Submission history

Useful for verifying if a drug is FDA-approved and when.

Inputs: api_key, application_number, approval_year, drug, +2 more. Returns: result

openfda_device_getterA

Get detailed information for a specific FDA device event report.

Retrieves complete device event details including:
- Device identification and specifications
- Complete event narrative
- Patient outcomes and impacts
- Manufacturer analysis and actions
- Remedial actions taken

Inputs: mdr_report_key (required), api_key. Returns: result

openfda_device_searcherA

Search FDA device adverse event reports (MAUDE) for medical device issues.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your research strategy!

Searches FDA's device adverse event database for:
- Device malfunctions and failures
- Patient injuries related to devices
- Genomic test and diagnostic device issues

By default, filters to genomic/diagnostic devices relevant to precision medicine.
Set genomics_only=False to search all medical devices.

Inputs: api_key, device, genomics_only, limit, +4 more. Returns: result

openfda_label_getterA

Get complete FDA drug label information by set ID.

Retrieves the full prescribing information including:
- Complete indications and usage text
- Detailed dosing instructions
- All warnings and precautions
- Clinical pharmacology and studies
- Manufacturing and storage information

Specify sections to retrieve specific parts, or leave empty for default key sections.

Inputs: set_id (required), api_key, sections. Returns: result

openfda_label_searcherA

Search FDA drug product labels (SPL) for prescribing information.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your research strategy!

Searches official FDA drug labels for:
- Approved indications and usage
- Dosage and administration guidelines
- Contraindications and warnings
- Drug interactions and adverse reactions
- Special population considerations

Label sections include: indications, dosage, contraindications, warnings,
adverse, interactions, pregnancy, pediatric, geriatric, overdose

Inputs: api_key, boxed_warning, indication, limit, +3 more. Returns: result

openfda_recall_getterA

Get detailed FDA drug recall information for a specific recall.

Returns complete recall details including:
- Full product description and code information
- Complete reason for recall
- Distribution pattern and locations
- Quantity of product recalled
- Firm information and actions taken
- Timeline of recall events

Inputs: recall_number (required), api_key. Returns: result

openfda_recall_searcherA

Search FDA drug recall records from the Enforcement database.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your research strategy!

Returns recall information including:
- Classification (Class I, II, or III)
- Recall reason and description
- Product identification
- Distribution information
- Recalling firm details
- Current status

Class I = most serious (death/serious harm)
Class II = moderate (temporary/reversible harm)
Class III = least serious (unlikely to cause harm)

Inputs: api_key, drug, limit, page, +4 more. Returns: result

openfda_shortage_getterA

Get detailed FDA drug shortage information for a specific drug.

Returns comprehensive shortage details including:
- Complete timeline of shortage
- Detailed reason for shortage
- All affected manufacturers
- Alternative products if available
- Resolution status and estimates
- Additional notes and recommendations

Data is updated periodically from FDA shortage database.

Inputs: drug (required), api_key. Returns: result

openfda_shortage_searcherA

Search FDA drug shortage records.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your research strategy!

Returns shortage information including:
- Current shortage status
- Shortage start and resolution dates
- Reason for shortage
- Therapeutic category
- Manufacturer information
- Estimated resolution timeline

Note: Shortage data is cached and updated periodically.
Check FDA.gov for most current information.

Inputs: api_key, drug, limit, page, +2 more. Returns: result

thinkA

REQUIRED FIRST STEP: Perform structured sequential thinking for ANY biomedical research task.

🚨 IMPORTANT: You MUST use this tool BEFORE any search or fetch operations when:
- Researching ANY biomedical topic (genes, diseases, variants, trials)
- Planning to use multiple BioMCP tools
- Answering questions that require analysis or synthesis
- Comparing information from different sources
- Making recommendations or drawing conclusions

⚠️ FAILURE TO USE THIS TOOL FIRST will result in:
- Incomplete or poorly structured analysis
- Missing important connections between data
- Suboptimal search strategies
- Overlooked critical information

Sequential thinking ensures you:
1. Fully understand the research question
2. Plan an optimal search strategy
3. Identify all relevant data sources
4. Structure your analysis properly
5. Deliver comprehensive, well-reasoned results

## Usage Pattern:
1. Start with thoughtNumber=1 to initiate analysis
2. Progress through numbered thoughts sequentially
3. Adjust totalThoughts estimate as understanding develops
4. Set nextThoughtNeeded=False only when analysis is complete

## Example:
```python
# Initial analysis
await think(
    thought="Breaking down the relationship between BRAF mutations and melanoma treatment resistance...",
    thoughtNumber=1,
    totalThoughts=5,
    nextThoughtNeeded=True
)

# Continue analysis
await think(
    thought="Examining specific BRAF V600E mutation mechanisms...",
    thoughtNumber=2,
    totalThoughts=5,
    nextThoughtNeeded=True
)

# Final thought
await think(
    thought="Synthesizing findings and proposing research directions...",
    thoughtNumber=5,
    totalThoughts=5,
    nextThoughtNeeded=False
)
```

## Important Notes:
- Each thought builds on previous ones within a session
- State is maintained throughout the MCP session
- Use thoughtful, detailed analysis in each step
- Revisions and branching are supported through the underlying implementation

Inputs: thought (required), thoughtNumber (required), totalThoughts (required), nextThoughtNeeded.

trial_getterA

Fetch comprehensive details for a specific clinical trial.

Retrieves all available information for a clinical trial by its NCT ID.
This includes protocol details, locations, outcomes, and references.

For specific sections only, use the specialized getter tools:
- trial_protocol_getter: Core protocol information
- trial_locations_getter: Site locations and contacts
- trial_outcomes_getter: Primary/secondary outcomes and results
- trial_references_getter: Publications and references

Inputs: nct_id (required). Returns: result

trial_locations_getterA

Fetch contact and location details for a clinical trial.

Retrieves all study locations including:
- Facility names and addresses
- Principal investigator information
- Contact details (when recruiting)
- Recruitment status by site

Useful for finding trials near specific locations or contacting study teams.

Inputs: nct_id (required). Returns: result

trial_outcomes_getterA

Fetch outcome measures and results for a clinical trial.

Retrieves detailed outcome information including:
- Primary outcome measures
- Secondary outcome measures
- Results data (if available)
- Adverse events (if reported)

Note: Results are only available for completed trials that have posted data.

Inputs: nct_id (required). Returns: result

trial_protocol_getterB

Fetch core protocol information for a clinical trial.

Retrieves essential protocol details including:
- Official title and brief summary
- Study status and sponsor information
- Study design (type, phase, allocation, masking)
- Eligibility criteria
- Primary completion date

Inputs: nct_id (required). Returns: result

trial_references_getterA

Fetch publications and references for a clinical trial.

Retrieves all linked publications including:
- Published results papers
- Background literature
- Protocol publications
- Related analyses

Includes PubMed IDs when available for easy cross-referencing.

Inputs: nct_id (required). Returns: result

trial_searcherA

Search ClinicalTrials.gov for clinical studies.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your research strategy!

Comprehensive search tool for finding clinical trials based on multiple criteria.
Supports filtering by conditions, interventions, location, phase, and eligibility.

Location search notes:
- Use either location term OR lat/long coordinates, not both
- For city-based searches, AI agents should geocode to lat/long first
- Distance parameter only works with lat/long coordinates

Returns a formatted list of matching trials with key details.

Inputs: age_group, conditions, distance, funder_type, +12 more. Returns: result

variant_getterA

Fetch comprehensive details for a specific genetic variant.

Retrieves all available information for a variant including:
- Gene location and consequences
- Population frequencies across databases
- Clinical significance from ClinVar
- Functional predictions
- External annotations (TCGA cancer data, conservation scores)

Accepts various ID formats:
- HGVS: NM_004333.4:c.1799T>A
- rsID: rs113488022
- MyVariant ID: chr7:g.140753336A>T

Inputs: variant_id (required), include_external. Returns: result

variant_searcherA

Search MyVariant.info for genetic variant DATABASE RECORDS.

⚠️ PREREQUISITE: Use the 'think' tool FIRST to plan your research strategy!

Important: This searches for variant DATABASE RECORDS (frequency, significance, etc.),
NOT articles about variants. For articles about variants, use article_searcher.

Searches the comprehensive variant database including:
- Population frequencies (gnomAD, 1000 Genomes, etc.)
- Clinical significance (ClinVar)
- Functional predictions (SIFT, PolyPhen, CADD)
- Gene and protein consequences

Search by various identifiers or filter by clinical/functional criteria.

Inputs: cadd_score_min, consequence, frequency_max, frequency_min, +12 more. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 35 tools

    Disambiguation3/5

    The tool set has clear domain separation (articles, trials, variants, etc.) but significant overlap exists between general and specialized tools. For example, 'fetch' overlaps with all the individual getters (article_getter, trial_getter, etc.), and 'search' overlaps with domain-specific searchers (article_searcher, trial_searcher, etc.). Descriptions help clarify, but agents may struggle to choose between overlapping tools.

    Naming Consistency4/5

    Most tools follow a consistent verb_noun pattern (e.g., article_getter, trial_searcher, variant_getter), with verbs like 'get', 'search', and 'fetch'. However, there are deviations: 'think' and 'fetch' are single words, and 'alphagenome_predictor' uses a noun_verb pattern. Overall, naming is mostly predictable but not perfectly uniform.

    Tool Count2/5

    With 35 tools, the count is excessive for a single server, leading to redundancy and complexity. Many tools could be consolidated (e.g., merging getters and searchers into fewer tools with parameters). This large number may overwhelm agents and increase the risk of misselection, especially given the overlapping functionalities.

    Completeness5/5

    The tool set provides comprehensive coverage across biomedical domains, including articles, trials, variants, genes, drugs, diseases, FDA data, and NCI resources. Each domain has both search and fetch capabilities, with detailed lifecycle support. No obvious gaps are present; the surface is well-rounded for biomedical research tasks.

    Maintenance

    ActivityActive
    ResponsivenessUsually handled

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      D
      maintenance
      A comprehensive Model Context Protocol server that enables advanced PubMed literature search, citation formatting, and research analysis through natural language interactions.
      12
      10
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      Integrates Taiwan-specific medical data including ICD-10 codes, FDA drug databases, and nutrition information into the Model Context Protocol. It enables AI models to query clinical guidelines, verify medical codes, and convert health data into FHIR R4 standardized formats.
      MIT
    • A
      license
      C
      quality
      C
      maintenance
      Evidence-grounded biomedical retrieval and summarization through the Model Context Protocol, enabling queries for biomedical evidence with citation-backed results.
      2
      MIT