Skip to content
Back to search

rbai-scholar

SkillWorks with Claude CodeGitHub

Search over 24 million biomedical journal abstracts (20M+ in PubMed) by meaning, exact terms or both at once.

Safety scan

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

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/insightwithhari--rbai-scholar/badge.svg)](/listings/insightwithhari--rbai-scholar)

HTML

<a href="/listings/insightwithhari--rbai-scholar"><img src="/api/listings/insightwithhari--rbai-scholar/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

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.

plugin.json · Manifest · 14 lines
  1. {
  2. "$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
  3. "name": "rbai-scholar",
  4. "version": "1.0.4",
  5. "description": "Search over 24 million biomedical journal abstracts (20M+ in PubMed) by meaning, exact terms or both at once.",
  6. "author": {
  7. "name": "The Dream Lab Innovations",
  8. "url": "https://thedreamlabinnovations.com"
  9. },
  10. "homepage": "https://thedreamlabinnovations.com/rbai-scholar",
  11. "repository": "https://github.com/Insightwithhari/rbai-scholar",
  12. "license": "MIT",
  13. "keywords": ["biomedical literature", "PubMed", "semantic search", "research"]
  14. }
README.md · README · 206 lines
  1. # RBAi Scholar
  2. **Find biomedical papers by what they mean, not only by the words they use.**
  3. Ask in plain language from ChatGPT, Claude, Cursor, VS Code or your own code, and get back real papers from
  4. over 24 million biomedical journal abstracts (more than 20 million of them in PubMed), each with its DOI, authors,
  5. journal, year, citation count and abstract.
  6. [Get a free key](https://thedreamlabinnovations.com/rbai-scholar#quickstart) ·
  7. [Website](https://thedreamlabinnovations.com/rbai-scholar) ·
  8. [Setup for every app](docs/mcp.md) · [REST API](docs/api.md) · [How we tested it](benchmark/README.md)
  9. ## Why use it
  10. - **Your words don't have to match the paper's.** Ask *"which protein stops neurons dying after a stroke"*
  11. and it finds papers that never use those words.
  12. - **Exact terms when you need them.** Gene symbols, variants and accessions (`TP53`, `p.Gly12Asp`, `rs334`),
  13. `"quoted phrases"`, `AND` / `OR` / `NOT`.
  14. - **Both at once.** Put the term that must appear in quotes inside a question, and only papers that contain
  15. it come back, ranked by meaning: *role of "MDM2" in stabilising mutant p53*.
  16. - **Filters:** year, journal, author, open access, citation count, language, and study type: trials, meta-analyses,
  17. reviews, guidelines, case reports, observational studies. Or look up one paper by its
  18. DOI, PubMed ID or title words.
  19. - **Real papers only.** Every result is a published paper with its DOI, and its PubMed ID when it has one, so your
  20. assistant points you to papers that exist.
  21. - **Check, expand, save.** Hand it a reference list and it says which PubMed IDs and DOIs are real papers (up to 50
  22. in one search). Give it one paper and it finds the papers closest to it. Ask which papers in RBAi Scholar it cites,
  23. or which of them cite it. Ask for the results as RIS or BibTeX and import them into Zotero, Mendeley or EndNote.
  24. - **Free to start:** 20 searches a day with a free key.
  25. ## Start in two minutes
  26. ### 1. Get a free key
  27. Enter your email at [thedreamlabinnovations.com/rbai-scholar](https://thedreamlabinnovations.com/rbai-scholar#quickstart),
  28. type the code we email you, and copy your key. It starts with `rbai_sk_` and works about a minute later.
  29. ChatGPT and Claude don't need a key: they sign you in with the same email code (step 2).
  30. ### 2. Add RBAi Scholar to your app
  31. **ChatGPT or Claude (website, desktop app or phone).** Add a custom connector with this address:
  32. ```
  33. https://scholar.thedreamlabinnovations.com/v1/mcp
  34. ```
  35. - **Claude:** Customize → Connectors → Add → Add custom connector. Paste the address, choose
  36. **Sign in when needed**, click Add, then **Connect**. The Free plan allows one custom connector.
  37. - **ChatGPT** (where your plan allows custom MCP servers): Plugins → + → Add custom MCP server. Paste the
  38. address, choose **OAuth**, then Create as a plugin.
  39. An RBAi Scholar page opens: enter your email, type the 6-digit code we email you, and click **Allow**. Your
  40. searches then use your own account's daily allowance (a new email gets a free account).
  41. Prefer a key? Put it at the end of the address (`…/v1/mcp?key=rbai_sk_your_key_here`) and choose **No sign-in**
  42. in Claude or **No authentication** in ChatGPT. Keep that address to yourself: anyone who has it can spend your
  43. searches.
  44. **Claude Code**
  45. ```bash
  46. claude mcp add --transport http rbai-scholar https://scholar.thedreamlabinnovations.com/v1/mcp \
  47. --header "x-api-key: rbai_sk_your_key_here"
  48. ```
  49. **Cursor, Kiro, Windsurf and most other apps** (`mcp.json`):
  50. ```json
  51. {
  52. "mcpServers": {
  53. "rbai-scholar": {
  54. "url": "https://scholar.thedreamlabinnovations.com/v1/mcp",
  55. "headers": { "x-api-key": "rbai_sk_your_key_here" }
  56. }
  57. }
  58. }
  59. ```
  60. VS Code, Claude Desktop's configuration file and other apps: [docs/mcp.md](docs/mcp.md).
  61. ### 3. Ask
  62. Your assistant decides when to search. Try:
  63. - *Find recent papers on how alpha-synuclein spreads between neurons in Parkinson's disease.*
  64. - *What has been found since 2020 about off-target effects of CRISPR base editors?*
  65. - *Papers by Jennifer Doudna on genome editing, 2020 onwards.*
  66. - *Open-access studies on the TP53 R273H mutation.*
  67. - *Highly cited papers on antibiotic resistance in Klebsiella pneumoniae.*
  68. - *Are these real papers: PMID 21376230, 10.1038/s41586-021-03819-2?*
  69. - *Find papers like PMID 21376230 from 2022 onwards, as a RIS file for Zotero.*
  70. - *Which papers cite PMID 21376230, and what does it cite?*
  71. - *Meta-analyses since 2020 on SGLT2 inhibitors in heart failure.*
  72. ## Free and Pro
  73. | | Searches | Price |
  74. |---|---|---|
  75. | No key (the website and the REST API) | 10 a day | free |
  76. | Free key | 20 a day | free |
  77. | Pro key | 500 a day | ₹49 a month or ₹490 a year |
  78. One key works in your apps and in the REST API, with one allowance for both. Only searches count: connecting an
  79. app and listing its tools need no key and never use the allowance. Allowances reset at 00:00 UTC (5:30 am in India).
  80. ## For programmers: the REST API
  81. ```bash
  82. curl -s https://scholar.thedreamlabinnovations.com/v1/corpus/search \
  83. -H 'Content-Type: application/json' \
  84. -d '{"query": "how do chaperones prevent protein aggregation", "topK": 3}'
  85. ```
  86. That works without a key, 10 searches a day. With a key, add `-H "x-api-key: rbai_sk_your_key_here"`. The
  87. answer is JSON: the ranked papers with their abstracts, and how your query was read.
  88. | Mode | Use it for | Example |
  89. |---|---|---|
  90. | `semantic` (default) | a question in plain language | `which protein stops neurons dying after a stroke` |
  91. | `keyword` | gene symbols, variants, accessions, exact phrases, `AND` / `OR` / `NOT` | `BRCA1 AND "homologous recombination" NOT mouse` |
  92. | hybrid: `semantic` with a term in quotes | a question where one gene, drug or method must be mentioned | `role of "MDM2" in stabilising mutant p53` |
  93. `POST /v1/corpus/papers` looks up to 50 PubMed IDs or DOIs, `POST /v1/corpus/similar` finds the papers closest to one
  94. paper, and `POST /v1/corpus/references` and `POST /v1/corpus/citing` list the papers in RBAi Scholar that a paper cites
  95. or that cite it; all four need a key. Every route takes `format` (`ris` or `bibtex` for a citation file), the searches
  96. and lists take `offset` for the next page, and `/search` and `/similar` take `studyType`.
  97. Every field, filter and error: [docs/api.md](docs/api.md). Python and Node.js examples, no packages needed:
  98. [examples/](examples).
  99. ## How well does it work?
  100. We tested it against PubMed, OpenAlex and NIH's LitSense 2.0. Every number can be recomputed from the files in
  101. this repository with one command.
  102. The tests ran on the corpus as it stood on 14 October 2026 (11.26 million records). The live service keeps
  103. adding papers (24.6 million records on 16 November 2026), so the same question asked later can bring back a
  104. slightly different list. On that day we re-sent the same questions to the live service: questions in your own
  105. words found their paper in the top 10 for 63.8% (65.0% on the tested version), questions with one required term
  106. for 93.3% (93.9%), and short identifier queries in keyword mode for 77.9% (79.1%):
  107. [benchmark/results/live-24m-2026-11-16.json](benchmark/results/live-24m-2026-11-16.json). The published numbers
  108. don't change with it: they are recomputed from the saved results in this repository, not from new searches.
  109. Results on newer versions of the corpus will be added as new, dated files next to these, never in their place.
  110. No single mode serves every need, so each was tested on the need it serves:
  111. - **Questions in your own words.** 500 questions were each written about one paper, the way a researcher would ask, without
  112. reusing the phrases of its abstract. The paper a question came from was in RBAi Scholar's top 10 for 65% of them, against 28% for
  113. OpenAlex (46% when we asked it again in November 2026, after it moved to the same embedding model), 19% for LitSense 2.0 and
  114. under 1% for PubMed, and the lead held on the 380 targets kept out of development.
  115. - **A question with one required term.** With the term marked in quotes, hybrid search found the paper in the
  116. top 10 for 94% of 328 questions, against 42% for LitSense 2.0 given the same term.
  117. - **Short exact queries** (a gene symbol, a compound, an acronym) are what keyword mode is for: it found the
  118. paper in the top 10 for 79% of 86 such queries, against 35% for meaning-based search.
  119. - **A first page worth reading.** In 4,389 blinded relevance grades, meaning-based search gave the most relevant
  120. first page, while hybrid search most often put the paper you have in mind first. A second blinded round of 5,752 grades,
  121. with OpenAlex's lists fetched again in November 2026, confirmed both. In that round our keyword mode's first page scored
  122. below OpenAlex's current keyword search (nDCG@10 0.384 against 0.449).
  123. - **Expert labels.** On RELISH, ordering candidates labelled by scientists, RBAi Scholar's ranking scored above
  124. MedCPT's article encoder (nDCG@10 0.843 against 0.796).
  125. Retrieval depth, the comparisons with MedCPT, the three index implementations and the statistics:
  126. [benchmark/README.md](benchmark/README.md). To recompute everything: `node benchmark/score.mjs`
  127. (Node.js 18 or newer, nothing to install).
  128. ## Good to know
  129. - It searches titles and abstracts, not full texts.
  130. - It finds papers; it does not judge them. Read the papers your assistant cites before you rely on them.
  131. ## When something goes wrong
  132. | You see | What to do |
  133. |---|---|
  134. | a message that searching needs you to sign in or a key | sign in (ChatGPT, Claude) or add your key (step 2) |
  135. | 403, `invalid_api_key` | a new key needs about a minute; after regenerating a key, use the new one |
  136. | "You've used today's 20 free RBAi Scholar searches" in an app, or 429 `quota_exceeded` from the REST API | the day's searches are used up; they come back at 00:00 UTC (5:30 am in India), or upgrade to Pro |
  137. More: [docs/mcp.md](docs/mcp.md#when-it-does-not-work).
  138. ## What's in this repository
  139. ```
  140. docs/api.md REST API reference: fields, filters, errors, limits
  141. docs/mcp.md setup for ChatGPT, Claude, Claude Code, Cursor, VS Code and others
  142. mcp/ those app configurations as files
  143. examples/ Python and Node.js search examples
  144. benchmark/ the evaluation: questions, every system's results, the grades, and the scorer
  145. server.json the MCP Registry entry
  146. plugin.json, mcp.json the Agent Plugins package that plugin directories such as cursor.directory read
  147. ```
  148. The search service's own code is not published. The benchmark contains no abstract text, because abstracts
  149. belong to their publishers and authors; it has DOIs, PubMed IDs, titles, years and journals instead.
  150. ## Cite and licence
  151. If you use RBAi Scholar or its benchmark in your work, please cite this repository: [CITATION.cff](CITATION.cff)
  152. (GitHub's "Cite this repository" button gives ready formats). A paper describing the benchmark is in preparation.
  153. Code (the scorer, examples and configurations) is MIT ([LICENSE](LICENSE)). The benchmark data, results, figures
  154. and documentation are CC BY 4.0 ([LICENSE-DATA.md](LICENSE-DATA.md)). Bibliographic metadata comes from OpenAlex
  155. (CC0) and PubMed. What changed and when: [CHANGELOG.md](CHANGELOG.md). To report a problem or suggest a change, see
  156. [CONTRIBUTING.md](CONTRIBUTING.md); for anything security-related, [SECURITY.md](SECURITY.md).
  157. Made by [The Dream Lab Innovations](https://thedreamlabinnovations.com).

Version history

  • v1.0.411/22/2026

Ratings and comments

Sign in to rate this skill.

Sign in

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