diff --git a/examples/foc_eliza/README.md b/examples/foc_eliza/README.md index 4e54c6c093..8dffcd384b 100644 --- a/examples/foc_eliza/README.md +++ b/examples/foc_eliza/README.md @@ -375,61 +375,60 @@ Execution info: event root: 0x6b25ec6fd717296603fb0818f24682d88bfdf9c55d995091865c582eed18dc71 ``` -### Why Store Characters and Memory On-Chain? +### Why On-Chain? -Currently, Eliza defines characters using static JSON files. By storing characters and memory on-chain, we unlock the following key advantages: +Storing AI characters and their memory on-chain provides three key advantages: 1. **Dynamic Evolution** - -On-chain storage allows the character’s `bio`, interests (`topics`), and message templates to evolve dynamically during conversations. - -This evolutionary process is fully transparent, enabling community participation and oversight of the character’s growth and changes. - -2. **Governance and Auditing** - -On-chain storage supports governance for defining character evolution rules. For example: - - - The community can audit and manage the AI Agent’s memory through on-chain governance to prevent memory contamination caused by user interactions. - - Public AI Agents can have their behavior audited to ensure alignment with community expectations. - -3. **Decentralized Trustworthiness** - -Storing characters and memory on-chain enhances Eliza’s trustworthiness, making it suitable for open and fair AI ecosystems. - -### Next Steps - -1. **Dynamic Character Loader** - -Enable Eliza to load character data from the blockchain dynamically and support real-time reloading of character definitions upon updates. - -2. **On-Chain Memory Sync Plugin** - -Develop a plugin to synchronize memory from the blockchain, allowing Eliza to update its state in real time during interactions. - -3. **AI Oracle Integration** - -Explore using an on-chain AI Oracle to provide decision-making capabilities for AI Agents, enhancing intelligent on-chain interactions. - -4. **Standardized On-Chain Components** - -Abstract the fundamental on-chain components of AI Agents and integrate them into the Move framework of Rooch to standardize on-chain AI Agent development. - -### How to participate - -1. **Push the Boundaries of Smart Contracts** - -Collaborate to explore the possibilities of implementing on-chain AI Agents, such as: - - - Supporting automatic state vectorization and vector retrieval directly in smart contracts(RGA on-chain). - - Developing more advanced on-chain AI inference mechanisms. - -2. **Provide More On-Chain Services** - -Build additional on-chain services tailored for AI Agents, such as: - - - Twitter/Blog services for AI Agents. - - On-chain chatroom services to support multi-user interactions with Agents. - -3. **Explore AI and DeFi/BTCFi Integration** - -Combine the dynamic capabilities of on-chain AI Agents with innovative financial applications. +- Characters evolve through conversations +- Bio, interests, and templates update dynamically +- Transparent growth process with community oversight + +2. **Governance & Auditing** +- Community-managed memory system +- Prevents memory contamination +- Public behavior auditing +- Alignment with community standards + +3. **Decentralized Trust** +- Enhanced trustworthiness +- Fair and open AI ecosystem +- Transparent character development + +### Roadmap + +1. **Character System** +- Dynamic character loading +- Real-time definition updates + +2. **Memory System** +- On-chain memory sync +- Real-time state updates +- Interaction history + +3. **AI Integration** +- On-chain AI Oracle +- Decision-making capabilities +- Smart contract inference + +4. **Framework Development** +- Standardized components +- Move framework integration +- Developer tools + +### Get Involved + +1. **Core Development** +- On-chain vector operations +- AI inference mechanisms +- Smart contract innovations + +2. **Service Building** +- AI Agent social services +- Multi-user chatrooms +- Character interaction systems + +3. **Integration** +- AI + DeFi applications +- BTCFi integration +- Cross-chain capabilities