Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.bland.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Node Level Testing

Validate dialogue prompts, variable extraction, and loop logic on individual pathway nodes with historical call data. Key Features:
  • Historical Call Testing: Select from past calls to test node prompts against real customer interactions
  • User Input Permutations: Generate variations of customer inputs using different communication styles to stress-test your prompts
  • Pinned Call Features: Select your most useful calls that automatically run with every test
Test individual nodes before deploying full pathways to catch issues early and ensure reliable customer experiences. Here’s an early demo of the feature.
Any questions or feedback? Feel free to reach out to us at hello@bland.ai

​
Improvements

Call Logs & Management
  • Added a call log mini player, removed the line overlay from the audio player for split human and agent audio, and improved key marker navigation.
Miniplayer interface for persistent audio playback
Citation & Analysis
  • [Enterprise] You can now run past calls against new citation variables during development to test citation quality
Citation testing setup interface
Web Widget
  • Added pathway logs to web widget dashboard
API & Development
  • [Enterprise] Enabled Node.js compatibility for Custom Code nodes
  • Fixed Bland Secrets to handle empty JSON bodies in POST requests for URL encoded refresh tokens