Skip to content

Conversation

ConstantTime
Copy link
Collaborator

@ConstantTime ConstantTime commented Oct 4, 2025

Summary

This PR adds support for configurable thinking indicator text, addressing issue #9885. Previously, the AI thinking indicator displayed a hardcoded "Thinking..." text during message generation. This enhancement allows admins to customize the indicator text.

Configuration Usage:

Users can now set custom thinking indicator text in their librechat.yaml:

interface:
  thinkingIndicatorText: "🤖 AI is processing..."  # Custom text shown while AI generates response

Change Type

  • New feature (non-breaking change which adds functionality)

Testing

Screenshot 2025-10-04 at 3 47 31 PM

Checklist

  • My code adheres to this project's style guidelines
  • I have performed a self-review of my own code
  • I have commented in any complex areas of my code (configuration is self-documenting)
  • I have made pertinent documentation changes (updated example config file)
  • My changes do not introduce new warnings (verified via linter check)
  • Local unit tests pass with my changes
  • Any changes dependent on mine have been merged and published in downstream modules (data-provider built successfully)

@ConstantTime ConstantTime changed the title Feat/issue 9885 feat: Add configurable thinking indicator text Oct 4, 2025
@ConstantTime ConstantTime marked this pull request as ready for review October 4, 2025 10:21
@ConstantTime ConstantTime requested a review from berry-13 October 4, 2025 10:23
- Add thinkingIndicatorText field to interface configuration schema
- Update ContentParts component to use configurable text instead of hardcoded translation
- Update example config with new option
- Maintain backward compatibility by falling back to default translation
- Add thinkingIndicatorText to interface schema in data-provider
- Update backend interface loader to pass through custom text
- Update ContentParts to use configurable text from startupConfig
- Fix optional @mcp-ui/client import in ToolCallInfo to prevent Vite errors
- Update librechat.example.yaml with configuration example
- Built packages and flushed cache
- Remove redundant comments and eslint disable
- Use destructured import with proper naming
- Simplify error handling with early return
@ConstantTime
Copy link
Collaborator Author

Hi @berry-13
FYI - This is similar as my other PR - #9937 and open for review too.

@ConstantTime
Copy link
Collaborator Author

Hi @danny-avila
Requesting a PR review for this.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant