https://github.com/daydreamsai/daydreams
Daydreams is a generative agent framework for executing anything onchain
https://github.com/daydreamsai/daydreams
agents ai
Last synced: 4 months ago
JSON representation
Daydreams is a generative agent framework for executing anything onchain
- Host: GitHub
- URL: https://github.com/daydreamsai/daydreams
- Owner: daydreamsai
- License: mit
- Created: 2024-12-20T22:17:20.000Z (6 months ago)
- Default Branch: main
- Last Pushed: 2025-02-17T09:06:39.000Z (4 months ago)
- Last Synced: 2025-02-17T10:22:46.428Z (4 months ago)
- Topics: agents, ai
- Language: TypeScript
- Homepage: https://docs.dreams.fun/
- Size: 54.2 MB
- Stars: 246
- Watchers: 18
- Forks: 66
- Open Issues: 22
-
Metadata Files:
- Readme: readme.md
- Changelog: changelog.md
Awesome Lists containing this project
- awesome-web3-ai - Daydreams
- awesome - daydreamsai/daydreams - Daydreams is a generative agent framework for executing anything onchain (TypeScript)
README
> ⚠️ **Warning**: This is alpha software under active development. Expect
> frequent breaking changes and bugs. The API is not yet stable.
# Cross-chain Generative Agents
[](https://opensource.org/licenses/MIT)
[](https://www.typescriptlang.org/)
[]()
[](https://docs.dreams.fun)
[](https://twitter.com/daydreamsagents)
[](https://github.com/daydreamsai/daydreams)
Daydreams is a powerful framework for building generative agents that can
execute tasks across any blockchain or API.
| Feature | Description |
| ---------------------- | -------------------------------------------------------------------- |
| 🔗 Chain Agnostic | Execute transactions and interact with any blockchain network |
| 👥 Multi-Expert System | Leverage specialized modules working together to solve complex tasks |
| 🧠 Context Management | Simple yet powerful memory and context handling system |
| 🎯 Goal-Oriented | Long-term planning and goal-oriented behavior capabilities |
| 💾 Persistent Memory | Built-in support for storing and retrieving long-term information |
| 🤔 Advanced Reasoning | Multi-step reasoning using Hierarchical Task Networks |
Want to contribute? Check our
[issues](https://github.com/daydreamsai/daydreams/issues) for tasks labeled
`good first issue`.
## Supported Chains
## Quick Start
### Prerequisites
- Node.js 18+ using [nvm](https://github.com/nvm-sh/nvm)
### LLM Keys
You'll need an API key for the LLM you want to use. We recommend using
[Groq](https://groq.com/) for most use cases.
- [OpenAI](https://openai.com/)
- [Anthropic](https://anthropic.com/)
- [Groq](https://groq.com/)
- [Gemini](https://deepmind.google/technologies/gemini/)
## Your First Dreams Agent
```bash
npm i @daydreamsai/core
```
Dreams agents are all functional. `createDreams` is a function that returns an
agent object, which can be run with `await agent.run()`. Inject discord,
telegram, or any other input/output to the agent and define your own actions.
```typescript
import { createGroq } from "@ai-sdk/groq";
import { createDreams, cli } from "@daydreamsai/core/v1";
// Initialize Groq client
const groq = createGroq({
apiKey: process.env.GROQ_API_KEY!,
});
// Create Dreams agent instance
const agent = createDreams({
model: groq("deepseek-r1-distill-llama-70b"),
extensions: [cli],
}).start();
```
Now chat via the CLI with the agent.
Read the [docs](https://docs.dreams.fun) for more information on how to use the
agent.
### Development
We use [bun](https://bun.sh/) for development.
```bash
bun install
```
## Contributing
Looking to contribute? We'd love your help!
If you are a developer and would like to contribute with code, please open an
issue to discuss before opening a Pull Request.
### Star History
[](https://star-history.com/#daydreamsai/daydreams&Date)