daydreams  by daydreamsai

Agent framework for onchain execution

created 7 months ago
485 stars

Top 64.2% on sourcepulse

GitHubView on GitHub
1 Expert Loves This Project
Project Summary

Daydreams is a lightweight, TypeScript-first framework for building stateful, autonomous AI agents. It targets developers seeking to create sophisticated, multi-contextual AI applications that can maintain memory across sessions and integrate with various LLM providers and platforms. The framework offers a flexible, modular architecture, enabling complex workflows and multi-step operations with a focus on developer experience and broad compatibility.

How It Works

Daydreams employs a modular, event-driven architecture centered around a "Dreams Engine" that orchestrates agent lifecycles. Key components include a Context Manager for handling multiple stateful conversations with type safety, a pluggable Memory System for persistent storage (KV and vector stores), and a type-safe Action Registry for extending agent capabilities. Agents can manage complex tasks via a Task Runner and integrate with external services through a flexible Extensions system, supporting various platforms and storage solutions.

Quick Start & Requirements

Highlighted Details

  • TypeScript-first with full type safety and IntelliSense.
  • Framework-agnostic: Integrates with LangChain, Vercel AI SDK, and custom LLM providers.
  • Multi-platform support via extensions (Discord, Twitter, Telegram, CLI).
  • Optional blockchain modules for Web3 interactions.

Maintenance & Community

Licensing & Compatibility

  • License: MIT.
  • Permissive license suitable for commercial use and integration into closed-source projects.

Limitations & Caveats

The framework is explicitly labeled as "Alpha Software," indicating that APIs may change and stability is not guaranteed. While it supports numerous LLM providers and platforms, extensive integration may require custom extension development.

Health Check
Last commit

21 hours ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
2
Star History
43 stars in the last 90 days

Explore Similar Projects

Starred by Wes McKinney Wes McKinney(Author of Pandas), Chip Huyen Chip Huyen(Author of AI Engineering, Designing Machine Learning Systems), and
9 more.

autogen by microsoft

0.6%
48k
Agentic framework for multi-agent AI applications
created 1 year ago
updated 1 day ago
Feedback? Help us improve.