Auto-GPT/autogpts/forge/tutorials/wip_005_adding_abilities.md

Ability Acquisition: Enhancing Your Agent's Capabilities

Introduction

• Understanding the Importance of Ability Acquisition
• The Concept of Abilities in AutoGPT

Section 1: Identifying Necessary Abilities

• Analyzing the Requirements for Your Agent
• Categorizing Abilities: Core vs. Supplementary

Section 2: Developing Abilities for Your Agent

• Integrating Existing Abilities from the Forge
• Developing Custom Abilities: A Step-by-step Guide

Section 3: Implementing and Executing Abilities

• Utilizing the Agent Protocol for Ability Implementation
• Executing Abilities: Task and Step Execution
• Example: Developing and Executing an Ability using Task and Step Schemas

Section 4: Encoding Abilities in Prompts for LLM Selection

• Understanding the Concept of Prompt Engineering
• Strategies for Effective Ability Encoding in Prompts
• Practical Examples: Encoding Various Abilities in Prompts

Section 5: Testing and Debugging Abilities

• Employing Test-Driven Development for Ability Testing
• Debugging Common Issues in Ability Implementation

Conclusion

• Recap of the Tutorial
• Preparing Your Agent for Ability Integration and Enhancement

Additional Resources

From The Rise and Potential of Large Language Model Based Agents: A Survey Zhiheng Xi (Fudan University) et al. arXiv. [paper] [code]

Research Papers

• [2023/07] ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs. Yujia Qin et al. arXiv. [paper] [code] [dataset]
• [2023/05] Large Language Models as Tool Makers. Tianle Cai et al. arXiv. [paper] [code]
• [2023/05] CREATOR: Disentangling Abstract and Concrete Reasonings of Large Language Models through Tool Creation. Cheng Qian et al. arXiv. [paper]
• [2023/04] Tool Learning with Foundation Models. Yujia Qin et al. arXiv. [paper] [code]
• [2023/04] ChemCrow: Augmenting large-language models with chemistry tools. Andres M Bran (Laboratory of Artificial Chemical Intelligence, ISIC, EPFL) et al. arXiv. [paper] [code]
• [2023/04] GeneGPT: Augmenting Large Language Models with Domain Tools for Improved Access to Biomedical Information. Qiao Jin, Yifan Yang, Qingyu Chen, Zhiyong Lu. arXiv. [paper] [code]
• [2023/04] OpenAGI: When LLM Meets Domain Experts. Yingqiang Ge et al. arXiv. [paper] [code]
• [2023/03] HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face. Yongliang Shen et al. arXiv. [paper] [code]
• [2023/03] Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models. Chenfei Wu et al. arXiv. [paper] [code]
• [2023/02] Augmented Language Models: a Survey. Grégoire Mialon et al. arXiv. [paper]
• [2023/02] Toolformer: Language Models Can Teach Themselves to Use Tools. Timo Schick et al. arXiv. [paper]
• [2022/05] TALM: Tool Augmented Language Models. Aaron Parisi et al. arXiv. [paper]
• [2022/05] MRKL Systems: A modular, neuro-symbolic architecture that combines large language models, external knowledge sources and discrete reasoning. Ehud Karpas et al. arXiv. [paper]
• [2022/04] Do As I Can, Not As I Say: Grounding Language in Robotic Affordances. Michael Ahn et al. arXiv. [paper]
• [2021/12] WebGPT: Browser-assisted question-answering with human feedback. Reiichiro Nakano et al. arXiv. [paper]
• [2021/07] Evaluating Large Language Models Trained on Code. Mark Chen et al. arXiv. [paper] [code]

Appendix

• Examples of Ability Implementations
• Glossary of Ability-Related Terms