Slava Kurilyak
Slava Kurilyak
When I run `doc index` on the [langchain](https://github.com/hwchase17/langchain) repository, I receive the following error: ```bash ⠇ Processing 494 files...Error during traversal: The text contains a special token that is not...
The **Recommendation Chain** can analyze the user’s goals, priorities, and previous task history to suggest specific tasks that need to be completed. This chain can be useful when asking the...
Inspired by agents with tool retrieval ([custom](https://python.langchain.com/en/latest/modules/agents/agents/custom_agent_with_tool_retrieval.html) and [babyagi](https://github.com/slavakurilyak/agentx/blob/main/agentx/baby_agi_with_tools.py)), we can create 'Agent Retrieval' which allows us to find the most relevant agent to accomplish user's task.
Inspired by [HuggingGPT](https://arxiv.org/pdf/2303.17580.pdf), we can create a task dependency chain, which parses the user request and determines the task dependency based on its knowledge. > Complex requests often involve multiple...
We can extend _babyagi_ to use a summarization chain that could assist with results summarization. This idea is inspired by @samwit's [Improve your BabyAGI with LangChain](https://youtu.be/DRgPyOXZ-oE?t=436).
Instead of waiting for LangChain to implement [Reflexion](https://github.com/hwchase17/langchain/issues/2316) or [RCI](https://github.com/hwchase17/langchain/issues/2646), we can implement a reflection chain with two prompts: > "Review your previous answer and find problems with your answer"...
We can explore role-playing agents, similar to [CAMEL](https://github.com/lightaime/camel), as suggested by @13331112522 on [Twitter](https://twitter.com/zhouql1978/status/1647009506457694208).
We can extend _babyagi_ to use [Koala](https://bair.berkeley.edu/blog/2023/04/03/koala/) (LLaMA-based model). This idea is inspired by @samwit's [Improve your BabyAGI with LangChain](https://youtu.be/DRgPyOXZ-oE?t=805).
Objective > Inspired by the several Auto-GPT related Projects (predominently BabyAGI) and the Paper "Generative Agents: Interactive Simulacra of Human Behavior", this python project uses OpenAI and Pinecone to Give...
Objective: > This Python script is an example of an AI-powered task management system. The system uses OpenAI and Pinecone APIs to create, prioritize, and execute tasks. The main idea...