Skip to content
Back to search

biomcp

BioMCP: Biomedical Model Context Protocol

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

Its scripts were only partly scanned.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/genomoncology/biomcp with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/genomoncology/biomcp'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/genomoncology--biomcp/badge.svg)](/listings/genomoncology--biomcp)

HTML

<a href="/listings/genomoncology--biomcp"><img src="/api/listings/genomoncology--biomcp/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

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

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 39 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "$schema": "https://code.claude.com/schemas/marketplace.json",
  3. "name": "biomcp",
  4. "owner": {
  5. "name": "GenomOncology",
  6. "email": "support@genomoncology.com"
  7. },
  8. "metadata": {
  9. "description": "BioMCP biomedical search and evidence retrieval for Claude Code and other agent harnesses."
  10. },
  11. "plugins": [
  12. {
  13. "name": "biomcp",
  14. "source": "./",
  15. "description": "Read-only biomedical search and evidence retrieval across BioMCP's CLI command grammar (genes, variants, trials, articles, drugs, diseases, pathways, proteins, and more). Wires the local BioMCP MCP server.",
  16. "version": "1.0.0",
  17. "author": {
  18. "name": "GenomOncology",
  19. "email": "support@genomoncology.com"
  20. },
  21. "homepage": "https://biomcp.org",
  22. "repository": "https://github.com/genomoncology/biomcp",
  23. "license": "MIT",
  24. "keywords": [
  25. "bioinformatics",
  26. "mcp",
  27. "genomics",
  28. "clinical-trials",
  29. "pubmed"
  30. ],
  31. "mcpServers": {
  32. "biomcp": {
  33. "command": "biomcp",
  34. "args": ["serve"]
  35. }
  36. }
  37. }
  38. ]
  39. }
README.md · README · 503 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # BioMCP
  2. One binary. One grammar. Evidence from the biomedical sources you already trust.
  3. ## Data terms
  4. 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](docs/reference/source-licensing.md) for the per-source breakdown.
  5. ## What is BioMCP?
  6. BioMCP is one CLI binary over a single command grammar that reaches 72 trusted
  7. biomedical sources directly (PubMed, ClinVar, ClinicalTrials.gov, OncoKB,
  8. Reactome, and more; ten further sources arrive inside another source's
  9. answer). It is also an MCP (Model Context Protocol) server, so the same tools are
  10. available to AI agents such as Claude Code, Codex, and Claude Desktop.
  11. BioMCP cuts through the usual biomedical data maze: one query reaches the
  12. sources that normally live behind different APIs, identifiers, and search
  13. habits. Researchers, clinicians, and agents use the same command grammar to
  14. search, focus, and pivot without rebuilding the workflow for each source. You
  15. get compact, evidence-oriented results across live public data plus local study analytics.
  16. ## Features
  17. - **Search the literature:** `search article` fans out across PubTator3 and
  18. Europe PMC and PubMed when the filters are compatible, deduplicates PMID/PMCID/DOI identifiers, and can add a Semantic
  19. Scholar leg when your filters support it.
  20. - **Pivot without rework:** move from a gene, variant, drug, disease, pathway,
  21. protein, or article straight into the next built-in view instead of
  22. rebuilding filters by hand.
  23. - **Choose a playbook:** `biomcp skill list` shows shipped worked examples
  24. so you can open the matching `biomcp skill <slug>` workflow.
  25. - **Analyze studies locally:** `study` commands cover local query, cohort, survival,
  26. compare, and co-occurrence workflows with native terminal, SVG, and PNG
  27. charts for downloaded cBioPortal-style datasets.
  28. - **Follow the paper trail:** `article citations`, `article references`,
  29. `article recommendations`, and `article entities` turn one known paper into a
  30. broader evidence map.
  31. - **Enrich and batch:** use `biomcp enrich` for top-level g:Profiler
  32. enrichment and `biomcp batch` for up to 10 focused `get` calls in one
  33. command.
  34. ## Quick start
  35. First useful query in under 30 seconds:
  36. ```bash
  37. uv tool install biomcp-cli
  38. biomcp health --apis-only
  39. biomcp skill list
  40. biomcp list gene
  41. biomcp search all --gene BRAF --disease melanoma # unified cross-entity discovery
  42. biomcp get gene BRAF pathways hpa
  43. ```
  44. ## Watch the introduction
  45. A project using BioMCP won St. Jude's KIDS BioHackathon in 2025. The talk that
  46. followed is a one-hour introduction to BioMCP, recorded October 24, 2025.
  47. [![BioMCP: An Introduction to Biomedical AI Agents](docs/assets/st-jude-2025-talk-1280x720.png)](https://www.youtube.com/watch?v=lXoe-4TENDE)
  48. [Watch on YouTube](https://www.youtube.com/watch?v=lXoe-4TENDE) &middot;
  49. [Read the write-up](https://biomcp.org/blog/st-jude-2025-hackathon-talk/) &middot;
  50. [Full transcript](https://www.imaurer.com/talks/biomcp-st-jude/)
  51. ## Installation
  52. ### Binary install
  53. ```bash
  54. curl -fsSL https://biomcp.org/install.sh | bash
  55. ```
  56. ### PyPI tool install
  57. ```bash
  58. uv tool install biomcp-cli
  59. # or: pip install biomcp-cli
  60. ```
  61. > **PyPI package warning:** install `biomcp-cli`, not `biomcp`. The `biomcp`
  62. > PyPI package is unrelated to this project.
  63. MCP Registry ownership marker: `mcp-name: io.github.genomoncology/biomcp`.
  64. This installs the `biomcp` binary in `~/.local/bin`. If that directory is not
  65. already on `PATH`, the installer prints one command to add it; it never edits
  66. your shell startup files.
  67. ### Homebrew
  68. ```bash
  69. brew tap genomoncology/biomcp
  70. brew install biomcp
  71. ```
  72. The separate `genomoncology/homebrew-biomcp` tap repository must exist before these commands can work.
  73. ### Docker
  74. ```bash
  75. docker run --rm ghcr.io/genomoncology/biomcp --version
  76. docker run --rm ghcr.io/genomoncology/biomcp list
  77. docker run --rm -i ghcr.io/genomoncology/biomcp serve
  78. ```
  79. Use the GHCR image for quick CLI checks or stdio MCP clients without a local install.
  80. ### Claude Code plugin
  81. Install the `biomcp` binary first, then add the hosted plugin marketplace and
  82. install the BioMCP plugin in Claude Code:
  83. ```text
  84. /plugin marketplace add genomoncology/biomcp
  85. /plugin install biomcp@biomcp
  86. ```
  87. The plugin wires Claude Code to the local stdio MCP server with `biomcp serve`.
  88. For guided BioMCP workflows, also install the skill assets below.
  89. ### Codex MCP server
  90. Install the `biomcp` binary first, then register the same stdio MCP server with
  91. Codex:
  92. ```bash
  93. codex mcp add biomcp -- biomcp serve
  94. ```
  95. ### Claude Desktop extension (.mcpb)
  96. Install BioMCP from the Anthropic Directory in Claude Desktop when that path is
  97. available for your environment. For local/manual setups, use the JSON MCP
  98. config below.
  99. ### Install skills
  100. The executable embeds the guide and worked examples. Read the complete guide
  101. with `biomcp skill render`, or install copies into your agent directory:
  102. ```bash
  103. biomcp skill install ~/.claude --force
  104. ```
  105. ### MCP clients
  106. ```json
  107. {
  108. "mcpServers": {
  109. "biomcp": {
  110. "command": "biomcp",
  111. "args": ["serve"]
  112. }
  113. }
  114. }
  115. ```
  116. ### Remote HTTP server
  117. For shared or remote deployments:
  118. ```bash
  119. biomcp serve-http --host 127.0.0.1 --port 8080
  120. ```
  121. Remote clients connect to `http://127.0.0.1:8080/mcp`. Probe routes are
  122. `GET /health`, `GET /readyz`, and `GET /`.
  123. Runnable demo:
  124. ```bash
  125. uv run --script examples/streamable-http/streamable_http_client.py
  126. ```
  127. See [Remote HTTP Server](https://biomcp.org/getting-started/remote-http/) for
  128. the newcomer guide.
  129. ### From source
  130. ```bash
  131. make install
  132. "$HOME/.local/bin/biomcp" --version
  133. ```
  134. For repo-local verification, run the standard gates directly: `make lint`,
  135. `make test`, and `make spec`. `make test` includes both Rust nextest and the
  136. Python/docs contract lane, while `make release-gate` adds the named full-feature
  137. proof and runs specs against the all-feature release binary.
  138. There is no supported `make check` command. Use `make verify` only for opt-in
  139. live public-upstream confidence; `make release-live-smoke` remains a
  140. compatibility alias.
  141. ## Command grammar
  142. ```text
  143. search <entity> [filters] → discovery
  144. skill list → playbook catalog for how-to questions
  145. discover <query> → concept resolution before entity selection
  146. get <entity> <id> [sections] → focused detail
  147. <entity> <helper> <id> → cross-entity pivots
  148. enrich <GENE1,GENE2,...> → gene-set enrichment
  149. batch <entity> <id1,id2,...> → parallel gets
  150. search all [slot filters] → counts-first cross-entity orientation
  151. ```
  152. ## Entities and sources
  153. The tables below distinguish detail-card entities from search-only surfaces so
  154. agents do not synthesize unsupported `get` commands.
  155. ### Gettable entities
  156. | Entity | Upstream providers used by BioMCP | Example |
  157. |--------|-----------------------------------|---------|
  158. | 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` |
  159. | variant | MyVariant.info, ClinVar, direct gnomAD v4 population data, CIViC, Cancer Genome Interpreter, OncoKB, cBioPortal, GWAS Catalog, AlphaGenome | `biomcp get variant "BRAF V600E" clinvar` |
  160. | 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` |
  161. | author | Semantic Scholar, ORCID | `biomcp get author semanticscholar:1716151` |
  162. | trial | ClinicalTrials.gov API v2, NCI CTS API | `biomcp search trial -c melanoma -s recruiting` |
  163. | diagnostic | NCBI Genetic Testing Registry local bulk bundle + WHO IVD local CSV + optional OpenFDA device overlay | `biomcp get diagnostic GTR000006692.3 regulatory` |
  164. | 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` |
  165. | 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` |
  166. | pathway | Reactome, KEGG, WikiPathways, g:Profiler, Enrichr-backed enrichment sections | `biomcp get pathway hsa05200 genes` |
  167. | protein | UniProt, InterPro, STRING, ComplexPortal, PDB, AlphaFold | `biomcp get protein P15056 complexes` |
  168. | adverse-event | OpenFDA FAERS/MAUDE/recalls plus CDC WONDER VAERS aggregate vaccine search | `biomcp search adverse-event --drug pembrolizumab` |
  169. | pgx | CPIC, PharmGKB | `biomcp get pgx CYP2D6 recommendations` |
  170. | cell-line | Cellosaurus with ChEMBL, PharmacoDB, DepMap, COSMIC, GDSC, and LINCS cross-references | `biomcp get cell-line CVCL_0004 xrefs` |
  171. ### Search-only entities
  172. | Entity | Upstream providers used by BioMCP | Example |
  173. |--------|-----------------------------------|---------|
  174. | gwas | GWAS Catalog | `biomcp search gwas --trait "type 2 diabetes"` |
  175. | phenotype | Monarch Initiative (HPO semantic similarity plus direct-support checks) | `biomcp search phenotype "HP:0001250"` |
  176. ## Cross-entity helpers
  177. Pivot between related entities without rebuilding filters.
  178. See the [cross-entity pivot guide](docs/how-to/cross-entity-pivots.md) for when
  179. to use a helper versus a fresh search.
  180. ```bash
  181. biomcp variant trials "BRAF V600E" --limit 5
  182. biomcp variant articles "BRAF V600E"
  183. biomcp drug adverse-events pembrolizumab
  184. biomcp drug trials pembrolizumab
  185. biomcp disease trials melanoma
  186. biomcp disease drugs melanoma
  187. biomcp disease articles "Lynch syndrome"
  188. biomcp gene trials BRAF
  189. biomcp gene drugs BRAF
  190. biomcp gene articles BRCA1
  191. biomcp gene pathways BRAF
  192. biomcp pathway drugs R-HSA-5673001
  193. biomcp pathway drugs hsa05200
  194. biomcp pathway articles R-HSA-5673001
  195. biomcp pathway trials R-HSA-5673001
  196. biomcp protein structures P15056
  197. biomcp article entities 22663011
  198. biomcp article citations 22663011 --limit 3
  199. biomcp article references 22663011 --limit 3
  200. biomcp article recommendations 22663011 --limit 3
  201. ```
  202. ## Gene-set enrichment
  203. ```bash
  204. biomcp enrich BRAF,KRAS,NRAS --limit 10
  205. ```
  206. Top-level `biomcp enrich` uses **g:Profiler**. Gene enrichment sections inside
  207. other entity views still reference **Enrichr** where that is the backing
  208. source.
  209. ## Sections and progressive disclosure
  210. Every `get` command supports selectable sections for focused output:
  211. ```bash
  212. biomcp get gene BRAF # summary card
  213. biomcp get gene BRAF pathways # add pathway section
  214. biomcp get gene BRCA1 diagnostics # diagnostic-test pivot from GTR
  215. biomcp get gene BRAF hpa # protein tissue expression + localization
  216. biomcp get gene BRAF civic interactions # multiple sections
  217. biomcp get gene BRAF all # standard sections; diagnostics/funding stay opt-in
  218. biomcp get variant "BRAF V600E" clinvar population conservation
  219. biomcp get article 22663011 tldr
  220. biomcp get drug pembrolizumab label targets civic approvals
  221. biomcp get drug trastuzumab regulatory --region who
  222. biomcp get disease "Lynch syndrome" genes phenotypes variants
  223. biomcp get disease tuberculosis diagnostics
  224. biomcp get diagnostic GTR000006692.3 regulatory
  225. biomcp get trial NCT02576665 eligibility locations outcomes
  226. ```
  227. In JSON mode, `get` responses expose `_meta.next_commands` for the next likely
  228. follow-ups and `_meta.section_sources` for section-level provenance. `batch ...
  229. --json` returns per-entity objects with the same metadata shape.
  230. ## API keys
  231. Most commands work without credentials. Optional keys improve rate limits or
  232. unlock optional enrichments:
  233. ```bash
  234. export NCBI_API_KEY="..." # ClinVar, PubTator, PubMed/efetch, PMC OA, NCBI ID converter
  235. export S2_API_KEY="..." # Optional Semantic Scholar auth; dedicated quota at 1 req/sec
  236. export OPENFDA_API_KEY="..." # OpenFDA rate limits
  237. export NCI_API_KEY="..." # NCI CTS trial search (--source nci)
  238. export ONCOKB_TOKEN="..." # OncoKB variant helper
  239. export ALPHAGENOME_API_KEY="..." # AlphaGenome variant effect prediction
  240. ```
  241. `search article`, `get article`, `batch article`, `get article ... tldr`, and
  242. the explicit Semantic Scholar helpers all work without `S2_API_KEY`. With the
  243. key, BioMCP sends authenticated requests and uses a dedicated rate limit at
  244. 1 req/sec. Without it, BioMCP uses the shared unauthenticated pool at 1 req/2sec.
  245. `search article --source` supports `all`, `pubtator`, `europepmc`, `pubmed`,
  246. `semanticscholar`, and `litsense2`. The default compatible article federation
  247. uses PubTator3, Europe PMC, PubMed, and automatic Semantic Scholar; use
  248. `--source semanticscholar` or `--source litsense2` explicitly when you want one
  249. of those sources alone. Explicit source selection also disables cross-provider
  250. row enrichment. References
  251. and recommendations can be empty for paywalled papers because of publisher
  252. elision in Semantic Scholar upstream coverage.
  253. ## Configuration
  254. ### Claude Desktop extension settings
  255. The directory bundle exposes only the optional settings needed for the first
  256. reviewer-facing build:
  257. | Claude Desktop field | Runtime env var | Purpose |
  258. |----------------------|-----------------|---------|
  259. | OncoKB Token | `ONCOKB_TOKEN` | Enables `biomcp variant oncokb "<gene> <variant>"` therapy and level evidence |
  260. | DisGeNET API Key | `DISGENET_API_KEY` | Enables scored DisGeNET sections on gene and disease lookups |
  261. | Semantic Scholar API Key | `S2_API_KEY` | Improves reliability for article TLDR, citation, reference, and recommendation helpers |
  262. The first directory build exposes only those three optional settings. Advanced
  263. CLI-only env vars remain documented in
  264. [API Keys](docs/getting-started/api-keys.md) for the general BioMCP CLI path.
  265. ## Usage Examples
  266. ### Public cross-entity overview
  267. **User prompt:** Give me a low-noise overview of BRAF in melanoma.
  268. **Expected tool call:** `biomcp search all --gene BRAF --disease melanoma --counts-only`
  269. **Expected behavior:** Returns a cross-entity counts summary that orients the
  270. next command instead of dumping long detail tables.
  271. **Expected output:** Counts-first summary with suggested next commands for the
  272. highest-yield entity follow-ups.
  273. ### Public variant evidence
  274. **User prompt:** Summarize ClinVar significance and population frequency for BRAF V600E.
  275. **Expected tool call:** `biomcp get variant "BRAF V600E" clinvar population`
  276. **Expected behavior:** Retrieves the focused variant card, ClinVar section, and
  277. population-frequency data in one read-only call.
  278. **Expected output:** Variant summary, ClinVar significance details, and gnomAD
  279. population frequencies.
  280. ### Credentialed OncoKB example
  281. **User prompt:** Show OncoKB therapy evidence for BRAF V600E.
  282. **Expected tool call:** `biomcp variant oncokb "BRAF V600E"`
  283. **Expected behavior:** Uses `ONCOKB_TOKEN` when configured and otherwise
  284. returns helpful guidance about the missing credential.
  285. **Expected output:** Therapy and level evidence when `ONCOKB_TOKEN` is set, or
  286. a clear setup hint when it is not.
  287. ### Credentialed DisGeNET example
  288. **User prompt:** Show scored DisGeNET associations for TP53.
  289. **Expected tool call:** `biomcp get gene TP53 disgenet`
  290. **Expected behavior:** Uses `DISGENET_API_KEY` to retrieve the scored
  291. gene-disease association section.
  292. **Expected output:** Ranked disease-association table with evidence counts and
  293. scores when `DISGENET_API_KEY` is configured.
  294. ## Privacy Policy
  295. BioMCP does not add telemetry, analytics, or remote log upload. Review the
  296. full privacy statement at <https://biomcp.org/policies/>.
  297. ## Multi-worker deployment
  298. BioMCP rate limiting is process-local. For many concurrent workers, run one shared
  299. Streamable HTTP `biomcp serve-http` endpoint so all workers share a single
  300. limiter budget:
  301. ```bash
  302. biomcp serve-http --host 0.0.0.0 --port 8080 \
  303. --allowed-hosts biomcp.example.org
  304. ```
  305. Loopback servers accept only local `Host` values by default. A non-loopback
  306. bind requires `--allowed-hosts`. The explicit `--unsafe-allow-any-host` escape
  307. hatch disables only this Host check; it does not add authentication, TLS, or
  308. encryption. Put remote deployments behind a trusted authenticated TLS proxy or
  309. inside a private network.
  310. Remote clients should connect to `http://<host>:8080/mcp`. Lightweight process
  311. probes are available at `GET /health`, `GET /readyz`, and `GET /`.
  312. ## Skills
  313. BioMCP ships an embedded agent guide and a worked-example catalog.
  314. Use `biomcp skill list` when you need the right worked example,
  315. then use `biomcp skill` to read the embedded BioMCP guide or install it into
  316. your agent directory when you want local copies of the workflow references:
  317. ```bash
  318. biomcp skill list
  319. biomcp skill
  320. biomcp skill install ~/.claude --force
  321. ```
  322. See [Skills](docs/getting-started/skills.md) for supported install targets,
  323. installed files, and legacy compatibility notes.
  324. ## Local study analytics
  325. `study` is BioMCP's local analysis family for downloaded cBioPortal-style datasets.
  326. The public entity surface handles API-backed, local-runtime, and hybrid
  327. discovery/detail; `study` commands work on local datasets when you need
  328. 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`.
  329. Use `study download` to fetch a dataset into your local study root. Set
  330. `BIOMCP_STUDY_DIR` when you want an explicit dataset location for reproducible
  331. scripts and demos; if it is unset, BioMCP falls back to its default study root.
  332. ```bash
  333. export BIOMCP_STUDY_DIR="$HOME/.local/share/biomcp/studies"
  334. biomcp study download msk_impact_2017
  335. 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
  336. biomcp study query --study msk_impact_2017 --gene RET --type sv
  337. ```
  338. See the [CLI reference](docs/user-guide/cli-reference.md#local-study-analytics)
  339. for the full `study` command family and dataset prerequisites.
  340. ## Ops
  341. ```bash
  342. biomcp version # show version and build info
  343. biomcp health # inspect API connectivity plus local DDInter/EMA/cache readiness
  344. biomcp update # self-update with release SHA256 checksum verification
  345. biomcp update --check # check for updates without installing
  346. biomcp uninstall # remove biomcp from ~/.local/bin
  347. ```
  348. ## Support
  349. - GitHub issues: <https://github.com/genomoncology/biomcp/issues>
  350. - Troubleshooting: [docs/troubleshooting.md](docs/troubleshooting.md)
  351. - Full documentation: <https://biomcp.org/>
  352. ## Documentation
  353. - [Getting Started](docs/getting-started/installation.md)
  354. - [Search All Workflow](docs/how-to/search-all-workflow.md)
  355. - [BioASQ Benchmark](docs/reference/bioasq-benchmark.md)
  356. - [Cross-Entity Pivot Guide](docs/how-to/cross-entity-pivots.md)
  357. - [Privacy Policy](docs/policies.md)
  358. - [Source Licensing and Terms](docs/reference/source-licensing.md)
  359. - [Data Sources](docs/reference/data-sources.md)
  360. - [Quick Reference](docs/reference/quick-reference.md)
  361. - [Troubleshooting](docs/troubleshooting.md)
  362. ## Citation
  363. If you use BioMCP in research, cite it via [`CITATION.cff`](CITATION.cff).
  364. GitHub also exposes `Cite this repository` in the repository sidebar when that file is present.
  365. ## Data Sources and Licensing
  366. 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.
  367. 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.
  368. Use [Source Licensing and Terms](docs/reference/source-licensing.md) for the per-source breakdown and [API Keys](docs/getting-started/api-keys.md) for setup steps and registration links.
  369. ## License
  370. MIT

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.