Free biomedical research and regulatory intelligence: FDA guidance, warning letters, drug labels, device approvals & scientific literature, with original-source citations.
Cursor is an AI-first code editor. It loads MCP servers from its settings and lets its agent call their tools as part of a task.
This guide connects Arca MCP to Cursor through MewCP, so Cursor can use Arca MCP's 4 tools, like search_medical, search_papers and fetch_source_record.
Arca MCP needs no credentials, but every call still runs through MewCP: scoped, checked and logged like any other.
More apps
Full Google Search Console control for search analytics, sitemaps, sites, and URL inspection, directly from your AI workflows. This MCP server exposes the Google Search Console API for querying search traffic data, managing sitemaps and site properties, and inspecting the indexing status of any URL. It is built for SEO teams monitoring search performance, developers automating sitemap and property management, and analysts pulling search traffic data into reporting pipelines.
Manage courses, assignments, rosters and student activity within Google Classroom. Create learning workflows, track submissions and organize educational content across classrooms.
PubMed enables AI agents to search biomedical and life sciences literature, retrieve research abstracts and publication details, filter studies by author, date, and topic, and explore scientific evidence to support medical research and literature reviews.
Official Ahrefs MCP server that lets AI agents securely access Ahrefs SEO data, site metrics, keyword research, and competitor analysis through the Ahrefs API.
Other clients
Arca MCP works the same way in every client below. Pick yours for its own guide.
Free to start.
Scope access to the exact Arca MCP tools and accounts you allow, work and personal side by side. Anything outside that scope is refused at call time, even when the model asks for it.
Each call is logged: what ran, on which account, and whether it worked. Failures come back structured, so the agent knows whether to retry instead of guessing.
Every app sits behind one secure MCP URL, and there is no server for you to deploy or keep running. Add an integration in the dashboard and it is live on the next call, with no config to change and no new credentials to wire up. Tools load only when a task needs them, so your agent stays fast and focused however many apps you connect.
A browser window opens the first time the client connects. Sign in and approve access. No API key needed.
Start a new chat and ask: "List the tools you can use from MewCP." You should see your connected apps.
Hybrid keyword + semantic search over Arca's mirrored public-source corpus, by source type: fda_records (FDA product records: labels and DailyMed, Drugs@FDA approvals and their approval packages (FDA's reviews, pharmacology/toxicology among them, and the administrative and correspondence documents holding FDA's letters and meeting minutes with the applicant), complete response letters (CRLs), Orange Book and Purple Book with their patents, NDC, shortages, FAERS, food, supplement, cosmetic and animal drug adverse events and tobacco problem reports, recalls and enforcement, 510(k)/De Novo/PMA decisions with their summaries, device classification, establishment registrations and device listings, MAUDE, AccessGUDID (UDI), postmarketing requirements, REMS, product-specific guidances, inactive ingredients, CBER biologics, warning and untitled letters, inspections, Form 483s, import refusals and alerts, orphan designations, advisory committee materials, Drug Trials Snapshots, safety communications, debarment lists, AI/ML devices, HDE approvals, substance identifiers (GSRS, UNII), RxNorm and ChEMBL); regulatory_guidance (FDA guidance, compliance manuals, compounding policy, FDA advisory committee meetings (briefing materials, minutes, transcripts), FDA press announcements, Federal Register notices (FDA, DEA, FTC) and Regulations.gov dockets with their comments, the CFR (21 CFR, HIPAA's 45 CFR, FTC's 16 CFR and other whole titles), CRLs, warning and untitled letters, product-specific guidances, the AI/ML-enabled device list, REMS, FTC cases and business guidance, HHS OCR guidance, US state law (CCHP telehealth policies, state health privacy statutes, California codes, state attorney general actions, pharmacy boards, NAAG, NABP, LegiScan bills), ad platform health policies, NRC Title 10 and ADAMS, and DOE's OSTI medical-isotope catalog); biomedical_literature (PubMed, SciELO, bioRxiv, medRxiv, arXiv and Europe PMC preprints (with open full text), NCBI Bookshelf books with chapter full text (GeneReviews, StatPearls and more), OpenAlex works PubMed does not index, and the reference data papers name: UniProt, PDB, MeSH, Open Targets and drug datasets (OFFSIDES, Drug Repurposing Hub)); clinical_trials (ClinicalTrials.gov (with posted protocols and SAPs, and AACT's day-by-day changes), EU CTIS, EudraCT, and national registries: ISRCTN, Health Canada, OMON, UMIN, DRKS, ReBEC, TCTR, PACTR, SLCTR, LBCTR, RPCEC; plus trial outcome labels from research datasets); public_health (CDC (data catalog, MMWR), ECDC publications, VAERS, CMS (Medicare Part B and D spending and prescribers, national and local coverage determinations and articles, HCPCS and OPPS pass-through codes, Open Payments, NPPES with NUCC taxonomy), Medicaid (NADAC, state drug utilization) and NIH RePORTER grants); clinical_guidelines (clinical practice guidelines and health technology assessments: HAS (France), AHRQ evidence reports, USPSTF recommendations, IQWiG reports, G-BA decisions (§35a benefit assessments among them) and SMC advice); standards_quality (ICH guidelines, PIC/S publications and FDA Recognized Consensus Standards); global_regulatory (non-US regulators: EMA (guidelines, medicine documents and EPARs, registers, Q&A), EU legislation and EudraLex, European Commission health guidance, EU MDR/IVDR expert panel opinions, EudraGMDP, MHRA, Health Canada (DPD, NOC, guidance, review documents, MDALL, recalls, Canada Vigilance), PMDA review reports, BfArM, PEI, AEMPS (CIMA), ANSM, AIFA, CBG-MEB, HSA and Medsafe); device_vigilance (field safety notices, recalls, adverse events and safety alerts for devices: FDA (MAUDE, recalls, enforcement, safety communications), BfArM, PMDA, Health Canada, ANVISA, ANSM and AEMPS); press_releases (company press releases from PR Newswire and Business Wire (PDUFA dates, trial readouts, approvals, deals) and FDA press announcements; searched only when named); company_filings (SEC EDGAR filings, Form D offerings and XBRL company facts; searched only when named); patents (US (USPTO) and European (EPO) patent grants and application publications with claims and description for drug, biologic and device classes, and PTAB trial and appeal decisions; searched only when named; credit European patent data as "This product contains data sourced from EPO databases, © European Patent Organisation"). Read-only lookup of already-published records: it cannot file, submit, or report anything to FDA, VAERS, MedWatch, or any other agency, so do not call it for requests to file or submit a report. When the question names a record type (warning letters, labels, adverse event reports, trials, recalls), pass the matching sourceTypes rather than searching the whole corpus. Results carry recordId for fetch_source_record and a citable source URL.
Ranked PubMed literature search: Arca's paper index fused with PubMed Best Match, with citation and open-access metadata from OpenAlex. Returns papers with title, authors, journal, year, DOI, PMID, abstract, citation count, and open-access status.
Read the indexed text of one public-source record by its recordId (returned by search_medical). Use after search when the excerpt is not enough. Returns a bounded text window; pass nextInput unchanged to continue. contentCapped reports text unavailable beyond the stored content.
Describe the corpus this server searches: the source lanes search_medical accepts in sourceTypes, with the underlying mirrored datasets in each.