Skip to content

ROADMAP.md

Latest commit

 

History

History
55 lines (46 loc) · 4.83 KB

ROADMAP.md

File metadata and controls

55 lines (46 loc) · 4.83 KB

InsightCoder Project Roadmap

This document outlines the planned development roadmap for the InsightCoder project. It is a living document and will be updated as the project evolves and we gather community feedback.

Short-Term Vision

  • Focus: Enhancing User Experience and Conversation Management.
    • Documentation Improvements:
      • Expand "Example Prompts" section in README.md with more diverse and practical examples.
      • Clarify in documentation: Add detailed explanation of conversation history, context window, and token usage in README.md to improve user understanding.
    • UI/UX Enhancements:
      • Conversation Folder Control:
        • High Priority: Implement a robust way for users to specify and manage the conversation folder, ensuring persistent conversation history even if the folder is Git-ignored. (e.g., command-line option, config file)
        • .gitignore Handling: Ensure conversation loading works correctly even if the conversation folder is in .gitignore (for privacy and to avoid Git tracking conversation history).
    • Automated Diff Application (Initial Stage):
      • Diff Detection: Implement logic to automatically detect code diff blocks within the AI's responses. (Look for Markdown code blocks with "diff" or similar language hints).
      • File Path Extraction & Validation: Extract the file path from the detected diff and validate that the file exists in the analyzed project.
      • High Priority & Crucial: Implement a pop-up confirmation dialog in the UI that appears when a diff is detected, asking the user "Would you like to apply these changes to {file_path} file?".
      • Visual Diff Display (Basic): Display the code diff in the confirmation dialog, ideally with a side-by-side or clear visual representation of the changes (similar to a simplified VS Code diff view).
      • Apply Diff on User Confirmation: If the user confirms, implement the functionality to automatically apply the detected diff to the specified file in the project.
    • Code Cleanup and Minor Refactoring:
      • Review code and add more comments, especially in chat_utils.py and ui.py.
      • Double-check and improve robustness of file path handling.

Medium-Term Goals

  • Focus: Introducing AI Code Agent Capabilities and Deeper Code Understanding.
    • Code Modification Suggestions (Initial Stage):
      • Experiment: Research and prototype the ability to suggest code refactoring opportunities based on AI analysis. (Initial focus: simple refactorings, e.g., function renaming, variable extraction).
      • UI Integration (Basic): Explore how to display refactoring suggestions in the UI (initially, perhaps as text output in the chat).
    • Improved Contextual Awareness:
      • Research: Investigate techniques for deeper code structure understanding (static analysis, code parsing libraries).
      • Symbol Understanding (Initial): Enhance AI's ability to understand basic code symbols (functions, classes) within a file.
    • Conversation History Management:
      • Implement a basic UI element to list saved conversations for easier access.
    • Automated Code Changes via Git Pushes (AI-Driven Development):
      • Feasibility Study & Design: Thoroughly research and design the architecture for automated code changes via Git pushes.
      • Git Integration: Implement Git functionality within InsightCoder to stage, commit, and push changes.
      • Authentication & Security: Securely manage GitHub credentials and implement robust security measures to prevent unintended pushes.
      • User Review Workflow: Design a workflow where AI-suggested changes are presented to the user for review and approval before Git push.
      • Automated Refactoring & Code Improvement: Develop AI capabilities to proactively identify and suggest code improvements, refactorings, and potentially even generate new code snippets, automatically creating Git commits for these suggestions.

Long-Term Vision

  • Focus: Evolving into a Proactive AI Code Agent, IDE Integration, and Advanced Features.
    • IDE/Editor Integration:
      • Explore and begin development of plugins/extensions for popular IDEs (VS Code, IntelliJ IDEA).
    • Advanced Features:
      • Project "Personality" and Customization (project-specific prompts, persona definition).
      • Continuous Code Monitoring and Real-time Insights.
      • Support for more LLMs.

This roadmap is subject to change and will be updated periodically based on progress, community feedback, and new opportunities.