User repository on GitHub
Scholars Reading List search API: semantic and keyword search over ~200M papers, map positions, citation walk
- Forks
- 4
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Tools & CLIs
Scholars Reading List · search
A search API over about 203 million OpenAlex English title and abstract records, running DenseOn embedding search by meaning and a tantivy index for exact words behind one HTTP service. Its whole-field topic maps use TypeSafe's Jev to judge archetype and cited papers.
- Stars
- 25
- Language
- Python
- License
- APACHE-2.0
- Reading time
- 2 min
- Published by
- Community
- Author
- MaximeRivest
- Last verified
- Oct 2026
- Languages
- Python
- Capabilities
- Open sourceSelf-hostedRequires API key
Highlights
- One HTTP API over ~203M English title+abstract records: DenseOn embeddings for meaning, a tantivy index for exact words, ~61 ms median end to end.
- Whole-field topic maps call TypeSafe's Jev to judge archetypes and works cited at least three times, tracked in fields/spend.json with a default $10/day budget.
- A private TypeSafe key lives in ~/.config/scholarsreadinglist/typesafe.env and is read only by the field-map worker; core search does not call Jev.
- An MCP server at /mcp exposes six read-only tools including search_papers, citation_walk and map_topic, plus literature_review and check_claim prompts.
- Apache-2.0 Python repo: 25 stars, 4 forks, created 2026-10-10; the README's intro still says no MCP server is built while a later section documents one.
Quickstart
bash
curl -G http://127.0.0.1:8810/search --data-urlencode "q=sleep apnea and atrial fibrillation" -d k=5Watch out
Grounded in the repository README fetched 2026-10-17; the digest excerpt truncates mid result-field list, so the Jev-powered whole-field map section and MCP server details come from the fuller README. The README contradicts itself about the MCP server, and the map accuracy numbers (89.5% recall, ~81% precision on one topic) are the project's own.
Questions about this entry
Short answers drawn from the linked project or page.
A service over roughly 203 million English title and abstract records from OpenAlex: search by meaning with DenseOn embeddings, exact-word search with a tantivy index, citation walk and topic maps, all behind one HTTP API.
In whole-field topic maps: srl_search/field.py asks TypeSafe's Jev to judge archetype papers, to label samples from near to far, and to judge works cited at least three times by the field.
No. The field worker reads TYPESAFE_API_KEY from ~/.config/scholarsreadinglist/typesafe.env, and the key is read by the worker only. Core search, the MCP tools and maps over the nearest 1,000 papers run without Jev.
The README measures one whole-field map — habitat fragmentation, 35,123 papers — at 13.5 minutes and $1.56, with fields/spend.json counting Jev dollars per day, and new fields stop when a daily budget (default $10) would be passed.
Query the API with curl, for example curl -G http://127.0.0.1:8810/search --data-urlencode "q=sleep apnea and atrial fibrillation" -d k=5. The public door at api.sciencemadereadable.com allows 30 requests per minute per address.
Six read-only tools — search_papers, get_papers, find_similar_papers, citation_walk, map_topic and read_topic_group — plus literature_review and check_claim prompts, connected with claude mcp add --transport http.
The digest covers only the first index (63.8M papers from 2024 onward), search times exclude 50-80 ms of question encoding, terse NFCorpus-style queries are the weak spot, and exact top-10 recovery tops out at 0.91 in the README's table.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: LinkedIn bookmarks search
X: 1,066 reaction videos, searchable
X: Intent-based search in a Chrome extension
X: Margin, bookmark search
More Jev resources like this
An Elasticsearch-compatible autoindex search layer that optionally sends top hits to TypeSafe Jev and reorders results by relevance scores.
Jev Tools & CLIs#community#rust#search
Community
xerj-org/xerjRust
xerj-org/xerjRustAn independent open-source pipeline that gives Jev vision: it detects and tracks objects in images, video, and RGB-D camera feeds, then asks Jev a typed question about each relevant object and prints the chosen answer with its confidence.
Jev Tools & CLIs#community#github#python
Community
CharlesFeng0314/JEV_seesPython
CharlesFeng0314/JEV_seesPythonA drop-in proxy that distills a repeated Jev classification call into a local model as it runs: it learns from Jev's own probability distributions, then answers most requests itself while uncertain ones and a 2% audit slice keep going to Jev.
Jev Tools & CLIs#community#python#distillation
Community
tomerglick57/JevstillerPython
tomerglick57/JevstillerPythonA local workbench that pairs OpenCV vision with Laya or Jev decision models: images are segmented and measured, hard rules settle the numbers, and the model gives the verdict — with one-click installers and an "open-source JEV" training path served at /v1/systemone.
Jev Tools & CLIs#community#github#python
Community
uu889/laya-opencvPython
uu889/laya-opencvPythonA Python batching proxy for Jev that returns Apache Arrow streams, with Python and JavaScript SDKs. It packs many independent states and questions into each TypeSafe API call or runs them concurrently, then emits the answers as an ordered Arrow IPC stream with the schema first.
Jev Repos & SDKs#community#github#python
Community
columnar-tech/jevaroPython
columnar-tech/jevaroPythonA Rotten Tomatoes-style ratings site for ML papers: reader votes plus a panel of 20 AI reviewer personas that scores every paper from day one, with Jev answering 20 typed questions per paper and classifying areas and paper types.
Jev Tools & CLIs#community#github#typescript
Community
LzyFischer/good-papersTypeScript
LzyFischer/good-papersTypeScriptFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.