4 posts tagged with "Academic Research"

Most "AI for research papers" demos stop at search: find a paper, summarize it, maybe extract a citation. Useful for a screenshot, useless for a real review.

Picture this instead. You have 240 papers saved in Mendeley for a RAG-evaluation review. Forty are missing DOIs. Eighteen have a citation record but no attached PDF. Six are duplicates from earlier exploratory searches. Your shared group library has 30 newer papers your collaborator added last week that you have not seen yet. None of that shows up in a "search the web" demo.

We rebuilt the Mendeley MCP server around that mess. An agent now works with your library as a library — saved papers, missing metadata, PDF files, folders, annotations, groups, trash, and all.

TL;DR​

• Query 586,479 registered studies on ClinicalTrials.gov live — not from a stale local mirror — through the Clinical Trials MCP server.
• Filter by condition, intervention, phase, recruiting status, sponsor, location, and posted results without building Lucene query strings; study detail returns eligibility, arms, outcomes, and site contacts in structured fields.
• Built for biotech landscape scans, clinical ops comparisons, patient-advocacy briefings, and research tools where the job is turn trial records into an answer, not navigate the registry one click at a time.

If you work in clinical research, biotech strategy, patient advocacy, or healthcare investing, the hard part is not knowing that ClinicalTrials.gov exists. The hard part is turning trial records into an answer you can use.

You may be trying to understand which sponsors are active in a disease area, whether a competitor has moved from phase 2 into phase 3, how strict the eligibility criteria are for a class of studies, or whether there are recruiting trials a patient advocacy team should know about. The raw registry has the data. Your actual job is to read across it quickly and explain what it means.

The Clinical Trials MCP server gives your AI agent a structured way to search studies, pull trial details, and summarize the result in the same conversation where the research question started.

The average MCP tool description is bad. Two recent arXiv papers measured exactly how bad. 97.1% of analyzed tools carry at least one description smell. 73% of servers reuse the same display_name across multiple tools. In head-to-head selection between five functionally equivalent servers, the one with a clearer description gets picked 72% of the time vs a 20% baseline.

A 260% selection lift from prose alone is the headline. Descriptions aren't documentation; they're part of the agent-facing prompt, and a bad one quietly costs you every selection round your server is in. So we built a scorer.

MCPBundles now runs an LLM-as-judge rubric over every published server, covering tool descriptions, server-level skill content, and the structured MCP annotations that the runtime actually reads. The verdict shows up on the public listing page.

arXiv hosts more than 2.4 million papers across physics, mathematics, computer science, quantitative biology, and more. Researchers, ML engineers, and academics increasingly want the same thing from their AI stack: search papers, read abstracts, follow citations, track new work in a field, and go deeper into full text — all from the same chat thread.

arXiv does not ship an official MCP server. Community implementations exist on GitHub, but they vary in quality, transport, and maintenance. MCPBundles hosts a dedicated arXiv provider (arxiv) with 4 MCP tools backed by the official arXiv Atom API — no API key required, no self-hosting, no config files. Enable the bundle, connect your AI client, and start searching.