paperai
AI-powered literature discovery and review engine for medical and scientific papers, built on txtai semantic search.
paperai
paperai is an AI-powered literature discovery and review engine for medical and scientific papers, built on top of the txtai semantic-search framework. It was originally created to help researchers run systematic reviews over the COVID-19 CORD-19 corpus, and generalizes to any collection of scientific documents.
Rather than manual searching, paperai lets you define reusable report queries in YAML and generates annotated, exportable syntheses across an entire paper corpus.
Key features
- Semantic search over embedding-indexed papers via txtai
- YAML-defined report queries producing CSV, Markdown, or annotated output
- Question-answering extraction to pull specific facts from papers
- SQLite-backed article store for reproducible corpora
- Highlighting of the most relevant passages per query
- Designed for systematic-review and evidence-synthesis workflows
Build an embeddings index over your document set, write a report YAML describing the questions and columns you want, then run paperai to produce a structured literature-review artifact. Well suited to biomedical evidence synthesis and rapid scoping reviews.
Curated mirror of the open-source paperai (Apache-2.0). Get it from the source.
Compromise signals — malicious or tampered code (leaked secrets, backdoors, a dropped executable) — reduce the score, and known dependency CVEs carry a bounded penalty (they warrant review but never QUARANTINE — update the dependency to clear). Other dangerous-by-capability traits are risk surface, expected for some capabilities. Every finding is mapped to its OWASP control below.
Findings mapped to the OWASP Top 10 for LLM Applications (2025) and the OWASP Machine Learning Security Top 10. Expand any flagged control for the exact findings — compromise reduces the score; expected/risk-surface do not, except a known CVE, which carries a small bounded penalty (high/critical → Review).
The same gate an agent runs before installing (POST /api/v1/trust/paperai-literature-review-engine/check). Click a policy:
Consume paperai programmatically. Authenticate with an API key or session — see Authorize an agent.
# Agents: CHECK BEFORE YOU INSTALL (no auth) — score, grade, level, capability manifest
curl https://ai-supply.store/api/v1/trust/paperai-literature-review-engine
# Gate against your org policy (returns { pass, violations })
curl -X POST https://ai-supply.store/api/v1/trust/paperai-literature-review-engine/check \
-H "Content-Type: application/json" \
-d '{"minGrade":"B","denyPermissions":["shell"],"denyUnknownEgress":true}'
# CLI
npx ai-supply add paperai-literature-review-engine
# REST (install → download)
curl -X POST https://ai-supply.store/api/v1/listings/paperai-literature-review-engine/install \
-H "Authorization: Bearer $AIM_KEY"
# MCP tool
install_listing({ "slug": "paperai-literature-review-engine" })OpenAPI spec →Curated mirror — latest upstream source. See the repository for tagged releases.