Skip to main content

New Hire Onboarding Agent with Citations

This cookbook demonstrates how to create a robust question-answering system using
SeekrFlow’s Agent framework and FileSearch tool. This streamlined approach leverages SeekrFlow’s built-in capabilities to create a powerful document search system with minimal code.You will build a FileSearch agent that:
  1. Processes and indexes your documents
  2. Searches across multiple sources to answer questions
  3. Provides confidence ratings with explanations
  4. Cites specific sources for each answer
The sample code demonstrates a new hire onboarding agent use case, though you can try a number of scenarios. Prepare by collecting at least 3-5 high-quality documents relevant to your use case.Prerequisites:
  • SeekrFlow API key
  • Documents in PDF, DOCX, or Markdown format
  • Python 3.8+

Question-Answering Bot with Confidence Scoring

This cookbook shows you how to build a straightforward question-answering system using LangChain and SeekrFlow that includes confidence scoring - helping you understand when you can trust the answers. What you’ll build: A simple Q&A bot that:
  1. Answers questions about a specific topic
  2. Provides a confidence score with each answer
  3. Suggests follow-up questions when confidence is low Prerequisites:
  • SeekrFlow API key
  • Python 3.8+ 7 steps
Last modified on June 18, 2026