# ElizaOS Weekly Newsletter **May 18 - May 24, 2025** ## Executive Summary This week marked significant progress for ElizaOS with the imminent arrival of v2 (expected to launch within days), along with substantial infrastructure improvements and plugin ecosystem enhancements. Multiple beta versions (.68 through .75) were released with rapid iterations addressing critical issues. The auto.fun platform is undergoing revitalization with plans for improved partner vetting and a more secure ecosystem. Meanwhile, the development team made major advances in Retrieval-Augmented Generation (RAG) functionality, which will greatly enhance agent knowledge capabilities. ## Development Updates ### Core Framework Enhancements The ElizaOS framework saw significant technical advancements this week: - **Knowledge System Overhaul**: The team factored Knowledge functionality out of the runtime into a dedicated plugin, creating a cleaner architecture. The new plugin-knowledge ([PR #4701](https://github.com/elizaos/eliza/pull/4701)) now supports PDF processing ([PR #4611](https://github.com/elizaos/eliza/pull/4611)), introducing robust document ingestion for agents. - **Database & API Improvements**: Database operations received substantial upgrades with an improved API ([PR #4556](https://github.com/elizaos/eliza/pull/4556)), enhancing entity retrieval capabilities and batch processing. New API endpoints were added for world and room management, enabling better multi-agent environments. - **Plugin Ecosystem**: The plugin publishing system was enhanced with NPM authentication and validation ([PR #4731](https://github.com/elizaos/eliza/pull/4731)), while plugin prefix validation was improved ([PR #4727](https://github.com/elizaos/eliza/pull/4727)). Several plugins were moved to dedicated repositories to streamline the core codebase. - **UI & UX Fixes**: Multiple fixes addressed UI issues in beta versions, particularly solving the critical problem of agents not appearing in the interface after updates to versions .68/.69, which required users to downgrade to .57 temporarily. ### Critical Bug Fixes - Fixed a critical "No handler found for delegate type: TEXT_EMBEDDING" error that was breaking agent functionality - Resolved Discord plugin compatibility issues with newer versions - Fixed knowledge plugin UUID syntax error and file processing for .md files - Addressed Twitter plugin Cloudflare authentication problems - Fixed CLI update issues with onnxruntime-node - Resolved PostgreSQL bypass and double initialization of server ## Community Spotlight The ElizaOS community has been actively helping each other navigate the latest beta releases and plugin issues: - **Jonas** has been particularly helpful in assisting users with Twitter plugin issues, recommending `plugin-twitter` instead of `agent-twitter-client` for better login functionality. - **0xbisbis** provided detailed assistance to users encountering Cloudflare authentication issues with Twitter integration, sharing instructions for setting up cookies with `TWITTER_COOKIES_AUTH_TOKEN`, `TWITTER_COOKIES_CT0`, and `TWITTER_COOKIES_GUEST_ID`. - **Ruby** offered comprehensive explanations of the auto.fun platform architecture, describing it as "GitHub but for autonomous systems" with secure agent-to-agent communication protocols and built-in safety features. - **Chinese community support** has been strengthened with **Kenk** creating and sharing a document with key updates at https://hackmd.io/@Kenkdoteth/S1giEQRWxg, while **Yikesawjeez** helped fix permissions issues for Chinese-speaking community members. ## Token Economics Discussions about the token ecosystem revealed important clarifications: - **ELI5 and Eddy**: These tokens will function as AI agents in the upcoming Eliza V2, to be part of "The Org" - but they remain community tokens, not official project tokens. - **Official Tokens**: Only AI16Z and DEGEN are confirmed as official tokens associated with elizaOS. - **Auto.fun Integration**: There's a growing focus on better integrating the token ecosystem with the auto.fun platform to incentivize legitimate long-term partnerships and discourage exploitative behavior. - **Liquidity Improvements**: Community members highlighted the need for improved liquidity for auto.fun platform tokens, with hopes that Eliza V2's launch will help revitalize the ecosystem. ## Coming Soon Several exciting developments are on the horizon: - **ElizaOS v2 Launch**: Expected within days, this major release will include new agents like Eli5 and Eddy, along with an improved agent terminal and expanded functionality. - **Partnercellerator Concept**: A more structured relationship with partners is being developed, starting with forum posts and eventually evolving to include staking contracts, dashboards, and proper vetting to prevent rug pulls. - **Enhanced RAG Capabilities**: The knowledge system is being improved with smarter semantic chunking that will respect markdown structure, enhancing agents' ability to work with complex documents. - **COM/AI16z Liquidity Pool**: The partnership with comput3.ai is proceeding, with plans for a liquidity pool expected to launch within the next two weeks. - **AI-Powered Survey System**: A new system for gathering stakeholder input using multiple-choice questions and a streak mechanism for daily participation is being developed. ## Resources - **GitHub Repository**: [elizaOS/eliza](https://github.com/elizaos/eliza) - **Documentation**: [docs.eliza.how](https://docs.eliza.how/) - **Discord Server**: [discord.gg/elizaOS](https://discord.gg/elizaOS) - **Twitter Plugin Setup**: [elizaos-plugins/plugin-twitter](https://github.com/elizaos-plugins/plugin-twitter) - **Chinese Community Updates**: [hackmd.io/@Kenkdoteth/S1giEQRWxg](https://hackmd.io/@Kenkdoteth/S1giEQRWxg) The ElizaOS ecosystem continues to evolve rapidly, with the much-anticipated v2 release promising to bring significant improvements to agent functionality, knowledge processing, and platform integration. The community remains actively engaged in supporting each other through these transitions while contributing to the ecosystem's growth and stability.