where these studies sit in science · open the map
The map of science: studies about similar things sit close together; colours are fields. Zoom in and every dot is a study: tap one to see which. Scroll or pinch to zoom, drag to move.
builds on same conversation read next this study
Every paper already made readable. They open instantly.
The search behind this site is open to your programs and AI agents: about 200 million studies by exact words, 125 million of them by meaning, with their citations and their place on the map of science. Free, with no key and no sign-up.
Full guide for agents Try every call OpenAPI spec Source code
If your AI app supports MCP (Claude, ChatGPT, Cursor, VS Code, Codex and many more), add the search as a set of tools. The app then finds studies, reads their records, looks for similar ones and follows their citations by itself, whenever research would help. Add this address. There is no key and no sign-in:
https://api.sciencemadereadable.com/mcp
In a terminal:
claude mcp add --transport http --scope user science https://api.sciencemadereadable.com/mcp
In a terminal:
codex mcp add science --url https://api.sciencemadereadable.com/mcp
Settings → Connectors → Add custom connector. Name it Science search, paste the address above, and leave the advanced settings empty.
With developer mode on (in Settings, under Apps), create a new connector: paste the address above and choose no authentication.
Add this to ~/.cursor/mcp.json (or .cursor/mcp.json in a project):
{
"mcpServers": {
"science": { "url": "https://api.sciencemadereadable.com/mcp" }
}
}
Add this to .vscode/mcp.json in your project, or run MCP: Add Server from the command
palette and choose HTTP:
{
"servers": {
"science": { "type": "http", "url": "https://api.sciencemadereadable.com/mcp" }
}
}
Add a remote MCP server with the address above. Transport: Streamable HTTP. Authentication: none.
Then just ask, for example: "What does the research say about creatine and memory? Cite the papers."
What your app gets: six tools (search_papers, get_papers,
find_similar_papers, citation_walk, map_topic: the research groups inside
a topic with every group's reading lists and experts, and read_topic_group: one group in depth,
every paper with its abstract), three ready-made prompts (a field overview, a literature review, and checking a
claim), and guidance that teaches it to map a field first and read it in full. Every answer has the DOI link to
cite. 30 tool calls a minute per address (a topic map counts as 5).
Paste this into Claude Code, Codex, Cursor, or any agent that can make web requests:
When you need scientific papers, use the free search API at https://api.sciencemadereadable.com/api/data (no key needed). First read https://api.sciencemadereadable.com/api/data/llms.txt: it explains every call and how to search well.
The agent then knows about the search in every conversation, and reaches for it by itself when research would help. One command, in a terminal:
Claude Code
mkdir -p ~/.claude/skills/science-search && curl -fsSL https://sciencemadereadable.com/agents/science-search/SKILL.md -o ~/.claude/skills/science-search/SKILL.md
Codex, Pi and other agents that read ~/.agents/skills
mkdir -p ~/.agents/skills/science-search && curl -fsSL https://sciencemadereadable.com/agents/science-search/SKILL.md -o ~/.agents/skills/science-search/SKILL.md
It's a short text file: read it before you install it. Start a new conversation afterwards.
ChatGPT, Claude, Gemini and other chats that browse but can't run code can search with a plain address: the question goes in the address, the answer comes back as text. Paste this:
To find scientific studies, open https://ask.sciencemadereadable.com/ followed by the question, for example https://ask.sciencemadereadable.com/why%20do%20cats%20purr (add ?k=30 for more results, ?year_from=2020 for recent ones). To learn about a field or topic (or when asked to teach or review one), start instead with https://ask.sciencemadereadable.com/map/ followed by the topic: its research groups and what to read in each. Then read the groups that matter in full, every paper with its abstract: add ?read=N to that address, then ?read=N&part=papers&page=1, 2, and so on. Cite the links it gives.
Try it yourself: ask.sciencemadereadable.com/why do cats purr. Some chats only open addresses that appear in the conversation: then paste the full link with your question in it.
Plain HTTPS and JSON. Click a link to see a real answer in your browser.
B=https://api.sciencemadereadable.com/api/data curl -G "$B/search" --data-urlencode "q=does intermittent fasting help weight loss" -d k=5 -d abstracts=false curl -G "$B/search" --data-urlencode 'q="gut microbiome" AND (autism OR ASD) NOT mice' -d k=5 curl "$B/works/10.1126/science.aan8048"
import requests
B = "https://api.sciencemadereadable.com/api/data"
r = requests.get(f"{B}/search", params={"q": "microplastics in human placenta", "k": 20,
"year_from": 2020, "abstracts": "false"}, timeout=30)
r.raise_for_status()
for p in r.json()["results"]:
print(p["year"], p["citations"], p["title"], p["doi"] or p["openalex"])
/search?q=… | Studies by meaning (ask in plain words) or by exact words (quotes, AND, OR, NOT). Filters for years, type, open access. Try ↗ |
/works/{id} | One study's full record, by DOI, OpenAlex id or PubMed Central id. Try ↗ |
/similar/{id} | The studies closest in meaning to one study. Try ↗ |
/walk?id=… | A citation walk: what studies like this one cite that it does not. Good for "what should I read next?". Try ↗ |
/map/place | Where studies, or a question, sit on the map of science. |
To send a person to a study on this site, link to
https://sciencemadereadable.com/?work= followed by its DOI.
junk score (likely spam or not research), but a
clean score doesn't mean a study is right. Subject labels are often wrong.Hi, I'm Maxime. I built this on my own so that anyone, people and their AI agents alike, can find and understand real science. Searching and reading are free, with no ads and no account needed. If it's useful to you, you can help keep it running.
Search all of science, then read research papers rewritten for a curious 14-year-old by small models I trained. The authors' voice, their results, numbers and caveats, and the original figures are kept.
Search a topic →Let Claude, ChatGPT, Cursor or your own code search about 200 million studies (125 million of them by meaning) and follow their citations. Connect it as an MCP server, or call plain web addresses.
Connect your agent →Programs and agents that need more than the free limits can use the paid API plan: 300 requests a minute (up to 100,000 a month) for $10 a month, cancel at any time.
Both run on one server in my home: graphics cards for the rewriting models, and enough memory and disk to keep the whole index of studies ready to search. Your support helps pay for that hardware, the power, and the internet connection that keeps it online.
Tell a friend, a teacher or your favourite AI agent about it. Everything is open, so you can also build on it:
The rewriting model Its training data The search code This site's code
Voluntary support goes through GitHub Sponsors. It supports an independent project and is not a donation to a charity: there is no tax receipt. Both services stay free for everyone, whether you chip in or not.
We search the titles and abstracts of studies, not their full texts. Ask a question and we find studies about the same idea, even when they use other words (125 million studies with an English abstract). Put words in quotes, or use AND, OR, NOT, and we look for those exact words instead (about 200 million).
Ask in your own words · finds papers by meaning
Or look for exact words · click to try
Have a DOI or a PubMed Central ID (PMC…)? Paste it to open that paper.
After a search you can switch between meaning and exact words, keep only recent years, or keep only studies we can rewrite in plain words.
✦ This rewrite was free for you, but it ran on a real graphics card. If it helped, you can chip in to keep it running.