{
  "server": "elizaOS Development",
  "title": "elizaOS Development Discord - 2025-01-17",
  "date": 1737072000,
  "stats": {
    "totalMessages": 157,
    "totalUsers": 40
  },
  "categories": [
    {
      "channelId": "1320246527268098048",
      "channelName": "💬｜general",
      "summary": "# Discord Chat Analysis\n\n## 1. Summary\nThe chat primarily revolves around technical issues and features of the Eliza framework. Users discussed the QuickIntel plugin API key acquisition process, with Mad requesting and receiving assistance from Ninja_Dev. Several deployment issues were addressed, particularly regarding server connectivity problems where the client was hardcoded to connect to localhost:3000, making remote deployment difficult. Leo sought help with modifying the backend connection address, and Santi identified the issue in the code. Users also inquired about the RAG (Retrieval-Augmented Generation) knowledge system, with questions about when it would be pushed to the main build and what file formats it supports (currently only TXT and MD files, not PDF). Other topics included implementing scheduled actions, Twitter Spaces functionality issues, and using tools with generateText. The chat demonstrates active community troubleshooting and developer support.\n\n## 2. FAQ\nQ: How do I modify the backend connection address from localhost:3000? (asked by Leo) A: Look for BASE_URL in client/src/lib/api.ts (answered by platodev and Santi)\nQ: Is PDF text extraction for knowledge RAG setup? (asked by 4n7m4n) A: No, currently only TXT or MD files. (answered by ꧁Ninja_Dev꧂)\nQ: Can we remove files from the knowledge section after they are inserted into the creation table? (asked by 4n7m4n) A: You can remove it. Once it's added it's in there. (answered by ꧁Ninja_Dev꧂)\nQ: Is there any example out there how to use rag or other features? (asked by platodev) A: Unanswered\nQ: How does finetuning work in eliza? (asked by platodev) A: You can try with the rag knowledge system and feed the agent with the data you need (answered by 0xn1c0)\nQ: What's the best way to make the agent run an action every 5 min? (asked by User) A: Unanswered\nQ: Does anyone know if the X Spaces feature is working? (asked by 0xn1c0) A: Unanswered\nQ: Is there some reference on how to correctly use tools with generateText? (asked by Blocknaut) A: Unanswered\n\n## 3. Help Interactions\nHelper: ꧁Ninja_Dev꧂ | Helpee: Mad | Context: Obtaining API key for QuickIntel plugin | Resolution: Directed Mad to check the profile section in the developer portal\nHelper: platodev | Helpee: Leo | Context: Modifying backend connection address from localhost | Resolution: Pointed to the BASE_URL in the codebase\nHelper: Santi | Helpee: Leo | Context: Hardcoded localhost connection issue | Resolution: Identified the issue and offered to create a PR\nHelper: 0xn1c0 | Helpee: platodev | Context: How to implement finetuning in Eliza | Resolution: Suggested using the RAG knowledge system to feed data to the agent\n\n## 4. Action Items\nTechnical: Fix hardcoded localhost:3000 in client/src/lib/api.ts | Description: Update connection URL to support remote deployment | Mentioned By: Santi\nTechnical: Enhance RAG system to remove entries from database | Description: Someone's working on functionality to remove entries from DB | Mentioned By: ꧁Ninja_Dev꧂\nTechnical: Debug X Spaces ICE connection failure | Description: Fix RTCPeerConnection error that crashes agent runtime | Mentioned By: 0xn1c0\nFeature: Add PDF support for knowledge RAG | Description: Currently only supports TXT and MD files | Mentioned By: 4n7m4n\nFeature: Implement scheduled actions for agents | Description: Allow agents to run actions at regular intervals (e.g., every 5 minutes) | Mentioned By: User\nDocumentation: Create examples for using RAG and other features | Description: Users need documentation on how to use these features | Mentioned By: platodev\nDocumentation: Provide reference for using tools with generateText | Description: Documentation needed for correctly using tools with models that support them | Mentioned By: Blocknaut\nDocumentation: Create documentation for Dexscreener plugin | Description: User unable to find documentation on how to get it running | Mentioned By: Mad",
      "messageCount": 72,
      "userCount": 18
    },
    {
      "channelId": "1324098367416172665",
      "channelName": "📮｜feedback",
      "summary": "# Analysis of Discord Channel \"📮｜feedback\"\n\n## 1. Summary\nThe chat segment contains feedback about the platform's documentation, technical challenges, and support issues. Users express frustration about information being scattered across videos rather than consolidated in documentation. There are specific concerns about understanding core system functionality, particularly regarding providers, plugins, and goals. One user mentions difficulty in getting technical support and responses to questions. Another user suggests using GitHub Copilot as a potential solution for finding information within repositories. The feedback highlights gaps in documentation, challenges for Python developers working with TypeScript, and a desire for better technical support and communication with the development team.\n\n## 2. FAQ\nQ: Would it be possible to get an agent to respond to Twitter DMs without the official Twitter API? (asked by nusko) A: Unanswered\nQ: Does using GitHub Copilot help with finding specific information in repositories? (asked by Display Name) A: Unanswered\n\n## 3. Help Interactions\nHelper: Display Name | Helpee: eric | Context: Difficulty finding information in documentation | Resolution: Suggested using GitHub Copilot to locate specific information in repositories, though no confirmation of success\n\n## 4. Action Items\nType: Documentation | Description: Improve documentation on plugins to reduce reliance on video tutorials | Mentioned By: eric\nType: Documentation | Description: Provide more information about the core system, especially how/when agents select items and providers | Mentioned By: eric\nType: Documentation | Description: Clarify how providers work, especially for generating information from actions | Mentioned By: eric\nType: Feature | Description: Implement workflow and multi-agent orchestration capabilities | Mentioned By: eric\nType: Documentation | Description: Create documentation on how to use goals | Mentioned By: eric\nType: Documentation | Description: Document how to make the system learn from environmental interactions | Mentioned By: eric\nType: Documentation | Description: Explain how to feed back content information resulting from actions | Mentioned By: eric\nType: Technical | Description: Reduce TypeScript boilerplate code needed for Python developers | Mentioned By: eric\nType: Technical | Description: Fix issues with the bridge | Mentioned By: Odilitime\nType: Feature | Description: Improve developer relations and technical support | Mentioned By: 0xn1c0",
      "messageCount": 5,
      "userCount": 5
    },
    {
      "channelId": "1324089429727514674",
      "channelName": "🤖｜agent-dev-school",
      "summary": "# Analysis of \"🤖｜agent-dev-school\" Discord Channel\n\n## 1. Summary\nThe channel discussions focus on ElizaOS development and integration issues. Users are working on various aspects including database management, knowledge integration, API connections (particularly with Telegram), and exploring core features like the goal evaluator. The framework appears to be in early development stages with some features not fully implemented yet. A user noted they would postpone using ElizaOS for workflows until it matures further, but appreciated its approach to agent-related problems. There are questions about character files, Instagram client integration, and interest in deferred execution functionality. Development resources were shared, including Chinese development guides for ElizaOS.\n\n## 2. FAQ\nQ: When you add new entries to knowledge, they get added to the character file though right? (asked by nusko) A: Unanswered\nQ: How to obtain the API token for ElizaOS to integrate with Telegram Bot? (asked by Alise) A: First google hit: https://communalytic.org/tutorials/obtaining-telegram-api-key/ (answered by tehsoul)\nQ: Why is working with default typescript character files not working? (asked by EsoTeros) A: Unanswered\nQ: Are there any plans for implementing deferred execution of actions? (asked by tem) A: Unanswered\nQ: How would I add the Instagram client from the develop branch to my package.json? (asked by Mihail Kirov) A: Unanswered\n\n## 3. Help Interactions\nHelper: fiction | Helpee: Unknown | Context: Database issue requiring reset | Resolution: Suggested deleting db.sqlite from agent directory and rerunning\nHelper: User | Helpee: Channel members | Context: Understanding goal evaluator functionality | Resolution: Shared GitHub links to relevant code and explained that goals are persistent but createGoal function isn't used, suggesting feature is under development\nHelper: m1hawk/风答 | Helpee: Channel members | Context: Need for ElizaOS development resources | Resolution: Shared YouTube playlist with Chinese development guides for ElizaOS\nHelper: tehsoul | Helpee: Alise | Context: Obtaining Telegram API token | Resolution: Provided link to tutorial for obtaining Telegram API key\n\n## 4. Action Items\nTechnical: Delete db.sqlite from agent directory and rerun to resolve database issues | Mentioned By: fiction\nTechnical: Implement deferred execution of actions for scheduling functionality | Mentioned By: tem\nTechnical: Add Instagram client from develop branch to package.json | Mentioned By: Mihail Kirov\nFeature: Implement workflows functionality | Mentioned By: eric\nFeature: Add agents group thinking capability | Mentioned By: eric\nFeature: Improve environment learning capabilities | Mentioned By: eric\nFeature: Enhance goal setting functionality | Mentioned By: eric\nFeature: Add Python integration | Mentioned By: eric\nFeature: Implement scheduling functionality | Mentioned By: eric",
      "messageCount": 13,
      "userCount": 11
    },
    {
      "channelId": "1323745969115893780",
      "channelName": "📥｜pull-requests",
      "summary": "# Discord Chat Analysis: 📥｜pull-requests\n\n## 1. Summary\nThe chat primarily revolves around pull request management for the ElizaOS project. Several developers are seeking reviews and merges for their contributions, including plugins for Twilio, Livepeer as an LLM provider, and other features. There's significant discussion about using AI tools to accelerate PR reviews, with AIFlow.ML expressing interest in using CodeRabbit and Codeium Forge to help with reviews. Odilitime confirms that CodeRabbit is available upon request. There's also mention of organizational tooling decisions, with Linear apparently being mandated for all projects despite earlier flexibility. AIFlow.ML mentions plans to add Nvidia inference capabilities and a plugin for nim during the weekend. The conversation demonstrates a collaborative environment where team members actively seek feedback, offer assistance with reviews, and discuss tooling improvements to streamline development workflows.\n\n## 2. FAQ\nQ: Can I use CodeRabbit or should I use my own license? (asked by AIFlow.ML @ ElizaOS) A: I think so, you just need to request a review on the PR. Look at some that Sayonara merged (answered by Odilitime)\nQ: Did you test with and without the supported image provider? (asked by AIFlow.ML @ ElizaOS) A: Yes (answered by tcm390)\nQ: Do we actually use Linear? (asked by yikesawjeez) A: I'm told the org made the call to force all projects to use it (answered by Odilitime)\nQ: Can I add coderabbit and test Forge to help speed up PR? (asked by AIFlow.ML @ ElizaOS) A: We have coderabbit, it's available on request (answered by Odilitime)\n\n## 3. Help Interactions\nHelper: tcm390 | Helpee: AIFlow.ML @ ElizaOS | Context: Checking if a PR fixed an issue | Resolution: AIFlow.ML reviewed and approved the PR after confirming it was tested properly\nHelper: AIFlow.ML @ ElizaOS | Helpee: Unknown user | Context: Fixed conflicts in PR #2434 | Resolution: Successfully resolved code conflicts\nHelper: tcm390 | Helpee: Ultros | Context: Twitter login failure issue | Resolution: Provided a link to a GitHub issue with relevant information\nHelper: Odilitime | Helpee: AIFlow.ML @ ElizaOS | Context: How to use CodeRabbit for PR reviews | Resolution: Explained that it's available on request and to look at examples from Sayonara\n\n## 4. Action Items\nType: Technical | Description: Review and merge Twilio plugin PR #2139 | Mentioned By: boolkeys\nType: Technical | Description: Review PR #2235 for inclusion in version 0.1.9 | Mentioned By: jonathan\nType: Technical | Description: Merge PR #2154 adding Livepeer as LLM provider | Mentioned By: DeFine\nType: Technical | Description: Review PR #2340 with incorporated feedback | Mentioned By: suns\nType: Technical | Description: Add Nvidia inference and plugin for nim | Mentioned By: AIFlow.ML @ ElizaOS\nType: Technical | Description: Fix conflicts in PR mentioned by Odilitime | Mentioned By: Odilitime\nType: Feature | Description: Create a Telegram <=> Linear bot for ticket data retrieval | Mentioned By: Odilitime\nType: Feature | Description: Develop a Discord plugin for Devin | Mentioned By: sayonara\nType: Technical | Description: Set up Linear for project management as mandated by organization | Mentioned By: Odilitime",
      "messageCount": 67,
      "userCount": 12
    }
  ]
}