Roadmap
Future plans and features for Daia.
Table of Contents
Overview
This page outlines planned features and improvements for Daia. If you’re interested in contributing to any of these features, check out the Development Guide and feel free to open an issue or pull request!
Core Features
Reply Message Rearrange
Status: Planned
Improve how Daia handles Discord reply chains and threaded conversations. This will allow the bot to better understand context when users reply to specific messages.
Benefits:
- Better context awareness in conversations
- Improved handling of threaded discussions
- More accurate responses to specific questions
Timezone Setting
Status: Planned
Allow users to configure timezone preferences for time-sensitive responses.
Benefits:
- Accurate time references in responses
- Scheduled features (future)
- Better localization support
daia-ignore Tag
Status: Planned
Add a way for users to mark messages that Daia should ignore, even in allowed channels.
Usage example:
[daia-ignore] This is a private conversation
Benefits:
- More control over bot interactions
- Privacy in shared channels
- Reduced noise in bot responses
Image Processing
Status: Planned
Enable Daia to analyze and respond to images shared in Discord.
Planned capabilities:
- Image description and analysis
- OCR (text extraction from images)
- Visual question answering
- Meme understanding
Technical approach:
- Leverage Gemini’s multimodal capabilities
- Support common image formats (PNG, JPG, GIF)
- Handle image attachments and URLs
YouTube Processing
Status: Planned
Allow Daia to understand and summarize YouTube videos.
Planned capabilities:
- Video summarization
- Transcript extraction
- Answer questions about video content
- Timestamp references
Usage example:
User: @Daia summarize this video: [YouTube URL]
Daia: This video discusses... [summary]
Deep Search
Status: Planned
Enhanced search capabilities beyond basic Google Search.
Planned features:
- Multi-source search aggregation
- Academic paper search
- Code repository search
- Fact-checking and source verification
Smart Reply Channels
Status: Planned
Intelligent channel detection that learns which channels need bot responses.
Features:
- Automatic detection of support channels
- Learning from user interactions
- Configurable per-server rules
- Activity-based response triggers
Gemini Safety Settings
Status: Planned
Configurable safety settings for Gemini API responses.
Configuration options:
- Harassment threshold
- Hate speech threshold
- Sexually explicit content threshold
- Dangerous content threshold
Usage:
# In .env
GEMINI_SAFETY_HARASSMENT=BLOCK_MEDIUM_AND_ABOVE
GEMINI_SAFETY_HATE_SPEECH=BLOCK_ONLY_HIGH
Extensibility
Mozilla any-llm Support
Status: Planned
Integration with Mozilla’s any-llm for broader LLM support.
Benefits:
- Support for multiple LLM providers
- Easier model switching
- Standardized interface
MCP Support
Status: Planned
Model Context Protocol support for enhanced tool use and context management.
Benefits:
- Better tool integration
- Improved context handling
- Standardized protocol support
OpenAI Support
Status: Planned
Native support for OpenAI models (GPT-4, GPT-3.5, etc.).
Configuration:
CHAT_MODEL_PROVIDER=openai
CHAT_MODEL=gpt-4
OPENAI_API_KEY=your_key_here
Benefits:
- More model choices
- Different pricing options
- Alternative capabilities
OpenAI Compatible Support
Status: Planned
Support for OpenAI-compatible APIs (LocalAI, Ollama, etc.).
Benefits:
- Self-hosted models
- Privacy-focused deployments
- Cost reduction for high-volume usage
Supported platforms:
- LocalAI
- Ollama
- LM Studio
- Text Generation WebUI
Database Support
Status: Planned
Persistent storage for conversation history and user preferences.
Planned databases:
- SQLite (simple deployments)
- PostgreSQL (production)
- MongoDB (flexible schema)
Features:
- Conversation history persistence
- User preferences storage
- Analytics and insights
- Cross-session context
Long-term Memory
Status: Planned
Enable Daia to remember information across sessions and conversations.
Capabilities:
- User preference learning
- Fact retention
- Relationship understanding
- Context accumulation over time
Technical approach:
- Vector database for semantic search
- Periodic memory consolidation
- Privacy-respecting storage
PocketFlow Deep Dive
Status: Planned
Comprehensive cookbook and documentation for PocketFlow integration.
Planned content:
- Advanced workflow patterns
- Custom node development
- Performance optimization
- Best practices guide
- Real-world examples
Community Requests
Have an idea for a feature? We’d love to hear it!
- Open an Issue: GitHub Issues
- Start a Discussion: GitHub Discussions
- Submit a PR: Check the Development Guide
Contributing
Want to help build these features? Here’s how:
- Pick a feature from the roadmap
- Open an issue to discuss your approach
- Fork the repository and create a feature branch
- Implement the feature following our coding standards
- Write tests for your implementation
- Submit a pull request to the
devbranch
See the Development Guide for detailed instructions.
Priority
Features are prioritized based on:
- Community demand: Features requested by users
- Impact: How much value the feature provides
- Complexity: Development effort required
- Dependencies: Prerequisites and technical requirements
Current high-priority items:
- Image Processing
- Database Support
- OpenAI Support
- Long-term Memory
Timeline
We don’t commit to specific timelines, as Daia is developed by volunteers. However, we aim to:
- Release minor updates monthly
- Release major features quarterly
- Maintain backward compatibility
- Keep documentation up to date
Feedback
Your feedback shapes Daia’s future! Let us know:
- Which features you’re most excited about
- What problems you’re trying to solve
- How you’re using Daia
- What could be improved
Join the conversation on GitHub Discussions.