DOI verification
How we verify DOIs via Crossref
The AutoSearch DOI validation methodology and why it matters for citation hallucination control.
Technical foundations
AutoSearch Research Blog
Methods notes on verified DOI citation, PRISMA-style automation, MCP agents, and internal RAG foundations for AI research assistants.
Comparison
AutoSearch vs ChatGPT for literature review
A technical comparison of general chat drafting and source-grounded literature-review automation.
PRISMA 2020
PRISMA 2020 automated methodology
How AutoSearch maps automated search, screening, retention, and disclosure to PRISMA 2020 concepts.
MCP
MCP servers for research assistants in 2026
Why Model Context Protocol matters for AI research agents and how AutoSearch exposes /mcp.
RAG
Internal RAG vs public-only research tools
Why proprietary indexed knowledge can improve research workflows when it is separated from public evidence.
EU MDR
MEDDEV 2.7/1 rev 4 literature search: a practical, auditable workflow
How to run a MEDDEV 2.7/1 rev 4 compliant literature search for clinical evaluation: PICO protocol, multiple databases, documented inclusion criteria, and a reproducible PRISMA flow.
PMCF
PMCF literature reviews: cadence, scope, and a repeatable workflow
Post-market clinical follow-up literature reviews under EU MDR: how often to re-run them, what to include, and how to keep the search strategy frozen and repeatable.
EU MDR
MDR Article 61: what 'systematic' actually requires from your literature review
An operational definition of 'systematic' for MDR Article 61 clinical evaluations: planned, reproducible, documented — mapped to PRISMA 2020 concepts notified bodies can audit.
Comparison
Elicit alternatives for systematic reviews (2026): an honest comparison
A criteria-based comparison of Elicit alternatives for systematic reviews: source coverage, citation verification, PRISMA support, output formats, and pricing.
Citation integrity
Fabricated citations in AI research tools: the data, and how to avoid them
What the 2026 evidence says about hallucinated references in AI research tools, and why verification by construction — every DOI resolved via Crossref — gets fabricated citations to zero.