5aea93d4c0
- Refactors the core `removeTask` function (`task-manager/remove-task.js`) to accept and iterate over comma-separated task/subtask IDs. - Updates dependency cleanup and file regeneration logic to run once after processing all specified IDs. - Adjusts the `remove-task` CLI command (`commands.js`) description and confirmation prompt to handle multiple IDs correctly. - Fixes a bug in the CLI confirmation prompt where task/subtask titles were not being displayed correctly. - Updates the `remove_task` MCP tool description to reflect the new multi-ID capability. This addresses the previously known issue where only the first ID in a comma-separated list was processed. Closes #140
49 lines
2.7 KiB
Plaintext
49 lines
2.7 KiB
Plaintext
# Task ID: 36
|
|
# Title: Add Ollama Support for AI Services as Claude Alternative
|
|
# Status: deferred
|
|
# Dependencies: None
|
|
# Priority: medium
|
|
# Description: Implement Ollama integration as an alternative to Claude for all main AI services, allowing users to run local language models instead of relying on cloud-based Claude API.
|
|
# Details:
|
|
This task involves creating a comprehensive Ollama integration that can replace Claude across all main AI services in the application. Implementation should include:
|
|
|
|
1. Create an OllamaService class that implements the same interface as the ClaudeService to ensure compatibility
|
|
2. Add configuration options to specify Ollama endpoint URL (default: http://localhost:11434)
|
|
3. Implement model selection functionality to allow users to choose which Ollama model to use (e.g., llama3, mistral, etc.)
|
|
4. Handle prompt formatting specific to Ollama models, ensuring proper system/user message separation
|
|
5. Implement proper error handling for cases where Ollama server is unavailable or returns errors
|
|
6. Add fallback mechanism to Claude when Ollama fails or isn't configured
|
|
7. Update the AI service factory to conditionally create either Claude or Ollama service based on configuration
|
|
8. Ensure token counting and rate limiting are appropriately handled for Ollama models
|
|
9. Add documentation for users explaining how to set up and use Ollama with the application
|
|
10. Optimize prompt templates specifically for Ollama models if needed
|
|
|
|
The implementation should be toggled through a configuration option (useOllama: true/false) and should maintain all existing functionality currently provided by Claude.
|
|
|
|
# Test Strategy:
|
|
Testing should verify that Ollama integration works correctly as a drop-in replacement for Claude:
|
|
|
|
1. Unit tests:
|
|
- Test OllamaService class methods in isolation with mocked responses
|
|
- Verify proper error handling when Ollama server is unavailable
|
|
- Test fallback mechanism to Claude when configured
|
|
|
|
2. Integration tests:
|
|
- Test with actual Ollama server running locally with at least two different models
|
|
- Verify all AI service functions work correctly with Ollama
|
|
- Compare outputs between Claude and Ollama for quality assessment
|
|
|
|
3. Configuration tests:
|
|
- Verify toggling between Claude and Ollama works as expected
|
|
- Test with various model configurations
|
|
|
|
4. Performance tests:
|
|
- Measure and compare response times between Claude and Ollama
|
|
- Test with different load scenarios
|
|
|
|
5. Manual testing:
|
|
- Verify all main AI features work correctly with Ollama
|
|
- Test edge cases like very long inputs or specialized tasks
|
|
|
|
Create a test document comparing output quality between Claude and various Ollama models to help users understand the tradeoffs.
|