What problem does it solve?
It solves the problem of finding and comparing experiment-level details that abstracts often omit, so you can quickly build evidence tables grounded in full-text results.
Core Features & Use Cases
- Full-text structured extraction: Retrieves 25+ fields per paper such as methods, results, sample sizes, quality scores, and conclusions rather than only titles and abstracts.
- Systematic review support: Helps with literature reviews and scoping studies by surfacing quantitative evidence and methodological differences across papers.
- Evidence synthesis readiness: Produces structured outputs suitable for comparing interventions, building evidence tables, and supporting meta-analytic or guideline workflows.
Quick Start
Configure the BGPT MCP server for your Claude/OpenAI-compatible MCP client, then ask: Search for papers about "CRISPR gene editing efficiency in human cells" and return their methods, sample sizes, results, and quality scores.