ARCHEMentor Evidence Tools
Why critical appraisal matters
Medical article appraisal AI agent

ARCHEMentor AI Based critical appraisal agent

Review medical literature faster with a guided appraisal workflow that extracts key study details, applies GRADE reasoning, and returns a practical verdict in plain language.

Input Types
DOI or article PDF
Output
Structured appraisal + verdict
Designed For
Clinicians, students, researchers

How to use this app

  • 1Choose your AI provider and model.
  • 2Enter your API key. If you use Ollama, you can leave the API key blank.
  • 3Upload the article PDF or type the DOI of the article.
  • 4Click Run Critical Appraisal and review the verdict, GRADE summary, and study details.

What the app will ask you for

  • AEnter your API key for OpenAI, Claude, Groq, Mistral, or Together AI.
  • BSelect a model from the dropdown so the app knows which provider to call.
  • CUpload the article PDF or enter the DOI to start the appraisal.
01 // LLM Configuration
Generate your API key from your provider dashboard: OpenAI, Claude / Anthropic, Groq, Mistral, Together AI.

How to use Ollama: install Ollama from docs.ollama.com, run `ollama serve`, then pull a local model such as `ollama pull llama3.2:3b`. In this app, choose Ollama (local), leave the API key blank, and select a local model.
02 // Article Input
Instructions:
  1. Enter your API key above.
  2. Choose a model provider and model.
  3. Upload the article PDF or type the DOI of the article.
  4. Press Run Critical Appraisal.
Use this when you know the paper DOI and want the app to fetch article text or abstract automatically.
Use a text-based PDF when possible. Very large or scanned PDFs may give weaker extraction quality.
Quick Guide
  • 1Enter your API key in the first field.
  • 2Select your provider from the dropdown.
  • 3Pick the model you want to use.
  • 4Type the DOI or upload the article PDF.
  • 5Read the verdict, reasons, applicability notes, and GRADE output.
What You Receive
  • VA practice-oriented verdict on whether the paper should change clinical thinking.
  • GA GRADE-based certainty summary with domain-level commentary.
  • SA concise structured summary of design, population, outcomes, and effect size.
Running appraisal. This may take 30 to 60 seconds depending on the provider and article length.
Verdict
03 // Appraisal
Why
    What Would Change This
      Applicability
        04 // GRADE Framework
        05 // Plain Language Summary
        06 // Study Details