Give Your AI Agent a Scientist's Library. a Science MCP Server

Valyu has launched a hosted science MCP server that gives AI agents access to full-text scientific literature, clinical trials, patents, and other structured data sources. The server exposes eleven tools, including valyu_search, valyu_academic_search, and valyu_bio_search, and can be integrated with Claude Desktop via a custom connector. A controlled before-and-after test demonstrates that scoping queries to specific scientific collections improves the relevance and depth of results.

Most "research agent" demos are searching abstracts and calling it literature review. The abstract tells you a Phase 3 melanoma immunotherapy trial hit its endpoint. It does not tell you the imaging protocol used to assess tumour response that lives in the methods section, or a supplementary table, or a figure caption. This walks through wiring a hosted science MCP server into your AI agent, scoping it to actual scientific collections, and running a controlled before-and-after test where exactly one variable changes. Valyu is a search API built for AI agents:one endpoint over biomedical literature, clinical trial registries, patents, financial filings and the open web, returning full text and structured metadata with resolvable identifiers rather than a list of links to go click. mcp-remote . query and max num results and nothing else. included sources in the API, then check result.source on every hit. include abstracts=True result.doi — read it, don't prompt for it. pip install valyu for the API examplesNo Node.js. The hosted server is remote HTTP. The endpoint is: https://mcp.valyu.ai/mcp?valyuApiKey=YOUR API KEY Auth rides in the query string. You can cap spend per session by appending &maxPrice=50 . Claude Desktop / claude.ai — go to claude.ai/settings/connectors https://claude.ai/settings/connectors → Add custom connector → paste the URL. Eleven tools, not one : | Tool | Arguments | Use | |---|---|---| valyu search | query , max num results , fast mode | Web search returning full page content | valyu academic search | query , max num results | Full text across arXiv, PubMed, bioRxiv, medRxiv | valyu bio search | query , max num results | PubMed, clinical trials, FDA labels, bioRxiv, medRxiv, ChEMBL, PubChem, DrugBank, Open Targets, NPI Registry, WHO ICD | valyu patents | query , max num results | Patent documents — claims, abstracts, inventors, filings | valyu contents | urls | Extract content from up to 10 URLs | valyu datasources | category | Enumerate the 36+ datasets at runtime | Plus valyu financial search , valyu sec search adds response length , valyu company research company , sections , valyu economics search and valyu datasources categories . API and SDK only — see the constraint above. Sources are addressed two ways through included sources : presets curated bundles and dataset IDs individual collections . Presets: academic , finance , patent , health , genomics , chemistry , physics , legal , politics , transportation , pulse , cybersecurity , environment , automotive , compliance , medical . Watch the preset boundaries — this bites people. academic covers literature and preprints only: | Dataset ID | Preset | Coverage | |---|---|---| valyu/valyu-pubmed | academic | 37M+ open-access biomedical papers, monthly | valyu/valyu-arxiv | academic | Physics, CS, maths, quant finance, economics | valyu/valyu-biorxiv | academic | 250K+ life-sciences preprints | valyu/valyu-medrxiv | academic | 80K+ clinical/health preprints | valyu/valyu-chemrxiv | academic | 30K+ chemistry preprints | valyu/valyu-clinical-trials | health | 500K+ ClinicalTrials.gov studies, real-time | valyu/valyu-drug-labels | health | 150K+ FDA labels via DailyMed | valyu/valyu-patents | patent | 8M+ USPTO filings, full text and figures | valyu/valyu-patents-epo | patent | 4M+ European filings from 1978 | valyu/valyu-chembl | chemistry | 2.5M+ bioactive compounds | valyu/valyu-pubchem | chemistry | 100M+ compounds | valyu/valyu-open-targets | chemistry | 60K+ drug targets | valyu bio search additionally reaches DrugBank, the NPI Registry and WHO ICD codes, which aren't broken out as dataset IDs in the datasources guide. Clinical trials are not in academic . If you scope a trial question to the academic preset you will get papers about trials, not registry records. Use health , or name valyu/valyu-clinical-trials directly. python from valyu import Valyu valyu = Valyu api key="YOUR API KEY" or set VALYU API KEY response = valyu.search "Phase 3 melanoma immunotherapy trials", search type="proprietary", all | web | proprietary | news included sources= "valyu/valyu-pubmed" , max num results=10, for result in response.results: print result.title print result.source check this print result.doi print result.content Then check the results. Every SearchResult carries a source field. After each search, confirm each result came from a collection you declared. If something arrives from elsewhere, treat the output as unscoped and rerun tighter. Filtering narrows the search; it is not a guarantee of exclusion. Note excluded sources accepts dataset IDs and domains but not presets. Here's the part worth running, with one honest caveat up front. By default include abstracts=False , PubMed search is restricted to papers that have available full text . Setting include abstracts=True expands the search to PubMed's complete abstract corpus and returns document-level abstracts. So this is not a clean single-variable A/B. Two things change at once: the corpus gets bigger, and the returned granularity drops to abstract level. It's still the sharpest comparison the API gives you, but describe it accurately — you are comparing full-text-only retrieval against broad abstract-level retrieval , not "the same search with and without full text." You want a detail that lives in the methods, a figure caption, or a supplement: What imaging protocol did the trial use for tumour response assessment? abstract run = valyu.search "Phase 3 melanoma immunotherapy tumour response assessment imaging protocol", search type="proprietary", included sources= "valyu/valyu-pubmed" , include abstracts=True, widen to the full PubMed abstract corpus max num results=10, This has to run in Python, not through the MCP client. No MCP tool accepts include abstracts — see the section above. Save the output verbatim. This is your baseline. fulltext run = valyu.search "Phase 3 melanoma immunotherapy tumour response assessment imaging protocol", search type="proprietary", included sources= "valyu/valyu-pubmed" , include abstracts=False, default — papers with available full text only max num results=10, Identical query, identical source, identical result count. Save that too. Put them side by side. Does the abstract-only answer contain the imaging protocol? Does the full-text one? If full text surfaces evidence abstract-only missed, you have a controlled result — for this query, this index, and this date. Things not to do with it: Four things, or nobody can reproduce it: the exact call parameters, the full verbatim output, the source list, and the correction. { "query": "<identical query string used in both runs ", "shared params": { "search type": "proprietary", "included sources": "valyu/valyu-pubmed" , "max num results": 10 }, "run abstract only": { "include abstracts": true, "output": "<full output, verbatim ", "sources returned": "<result.source values " }, "run full text": { "include abstracts": false, "output": "<full output, verbatim ", "sources returned": "<result.source values " }, "run date": "<YYYY-MM-DD ", "correction": { "missed by abstract": "<what was missing ", "found in full text": "<what full-text surfaced ", "source doi": "<result.doi ", "location": "<methods / figure caption / supplement " } } Record the date — PubMed syncs monthly and trials update in real time, so the same call will drift. Every claim links to a resolvable identifier. This is the line between a science agent and a chatbot with a search tool. | Source type | Identifier | Resolves at | |---|---|---| | Journal articles | DOI | https://doi.org/<doi | | Preprints bioRxiv, medRxiv, ChemRxiv | DOI | https://doi.org/<doi | | Clinical trials | NCT number | https://clinicaltrials.gov/study/<nct | | US patents | USPTO patent number | USPTO patent search portal | You don't have to parse these out of prose — SearchResult exposes doi , citation , authors , publication date , citation count and source as structured fields. Read them directly rather than asking the model to extract them. Do not demand a DOI for everything. Trials and patents have their own registries, and a prompt that insists on DOIs produces fabricated ones. For every factual claim, cite a resolvable identifier: a DOI for journal articles, an NCT number for clinical trials, or a patent number for patents. If the result has no identifier, give the URL and state that the claim is unverified. Never construct an identifier that was not returned. Three scoped searches, correct preset for each: python from valyu import Valyu valyu = Valyu api key="YOUR API KEY" Literature — academic preset lit = valyu.search "PD-1 inhibitor combination therapy melanoma", search type="proprietary", included sources= "academic" , Clinical trials — registry records live in health, NOT academic trials = valyu.search "PD-1 inhibitor melanoma Phase 3", search type="proprietary", included sources= "valyu/valyu-clinical-trials" , Patents — USPTO full text and figures patents = valyu.search "PD-1 antibody immunotherapy", search type="proprietary", included sources= "valyu/valyu-patents" , DOIs for the literature, NCT numbers for the trials, patent numbers for the patents. Valyu's DeepResearch POST /v1/deepresearch/tasks spans the same catalogue asynchronously. It can reach across domains in one task, but verify the returned sources match your intended scope before treating the output as complete. Source provenance — read result.source on every result. If it isn't a collection you declared, the run is unscoped. Identifier resolution — verify the cited identifier actually resolves before presenting the claim. DOI at doi.org, NCT at clinicaltrials.gov, patent through USPTO. Doesn't resolve → unverified. Full-text availability — PubMed full text is open access only, and include abstracts=True means you got abstracts instead of full text. Have the agent state which mode it ran in. An agent reasoning over an abstract as though it read the paper is the failure this whole post is about. Preprint status — bioRxiv, medRxiv and ChemRxiv are not peer-reviewed. Label them as preprints, with server name and DOI, so the reader can judge evidence level. Citation entailment — when the agent says a source supports a statement, confirm the passage is actually in the returned content . If it cites a figure, confirm the figure came back. Missing assets — figures, tables and supplements are not retrievable from every source. valyu/valyu-patents is the one dataset documented as carrying full text and figures; don't assume that generalises to the preprint servers. If an asset isn't there, the agent says so rather than substituting. Retrieval is not a redistribution licence. The Valyu Acceptable Use Policy https://www.valyu.ai/valyu-acceptable-use-policy applies across all APIs, datasets, models and indexes. You must not: The contents endpoint is your responsibility. Per the AUP: "You — not Valyu — are the party responsible for ensuring that your use of the Contents endpoint in connection with any given URL is lawful and authorised." Before submitting a URL, review the target's terms and acceptable use policy, and confirm automated extraction isn't prohibited by robots.txt , X-Robots-Tag headers or <meta name="robots" directives. In practice: the agent reads and reasons over retrieved content in-session and does not store it for redistribution. And retrieved research is not a substitute for professional medical advice, or for a human reading the primary source. included sources set, with the right preset for the source type max num results include abstracts differs between the two runs, and the writeup says the corpus widened too result.source checked on every result doi , NCT number or patent number /v1/contents without checking terms and robots directives How do I run the same question through both abstract-only and full-content workflows? In Python, not through MCP — no MCP tool exposes include abstracts . Issue the identical query twice against valyu/valyu-pubmed , once with include abstracts=True and once with the default False , keeping search type , included sources and max num results fixed. Save both outputs verbatim. Can I restrict my Claude Desktop agent to just PubMed? No. MCP search tools accept only query and max num results . Your scope control is which tool the agent picks. For real source pinning, call the API directly. What makes a good test question? One where the decisive detail sits outside the abstract — imaging protocols, assay conditions, eligibility subtleties. "What imaging protocol did the Phase 3 melanoma immunotherapy trial use for tumour response assessment?" works because that lives in methods or a supplement. Why did my clinical trial search return papers instead of registry records? You almost certainly scoped to the academic preset. Clinical trials live in health — use included sources= "valyu/valyu-clinical-trials" . How should the agent cite results without DOIs? NCT number for trials, patent number for patents. If none exists, the URL plus an explicit note that the claim is unverified. Read result.doi rather than having the model extract it.