feat: split langchain way and custom way; add description and docs for these two ways

[seancoding-day]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello @seancoding-day, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request significantly enhances the agent-based evaluation capabilities by formalizing two distinct implementation patterns: a framework-driven approach utilizing LangChain and a developer-driven custom workflow. This provides users with clear guidance and flexibility in building agents, whether they prefer rapid prototyping with LangChain's orchestration or fine-grained control over execution steps. The changes include a new LangChain-based fact-checking agent, updates to an existing custom workflow agent, and extensive documentation to support these new paradigms.

Highlights

• New Agent Implementation Patterns: Introduced two distinct agent implementation patterns: LangChain-Based (framework-driven) and Custom Workflow (developer-driven), providing flexibility for agent development.
• New LangChain-Based Agent: Added a new AgentFactCheck class, implementing a LangChain-based fact-checking agent with autonomous web search capabilities and structured output parsing.
• Updated Custom Workflow Agent: Refactored the AgentHallucination class to explicitly represent the Custom Workflow pattern, detailing its imperative control flow and updated configuration example.
• Enhanced Base Agent Functionality: Enhanced the BaseAgent class with a new _format_agent_input method, allowing subclasses to customize how input data (prompt, content, context) is prepared for LangChain agents.
• Comprehensive Documentation: Updated documentation (README.md, README_zh-CN.md, and docs/agent_development_guide.md) to comprehensively describe the new agent patterns, their use cases, implementation details, and a decision tree for selection.
• Dedicated Test Suite: Included a dedicated test suite for AgentFactCheck to ensure its functionality and robustness across various input and output scenarios.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request does a great job of splitting the agent implementation into two distinct patterns: a LangChain-driven approach and a custom workflow. The introduction of AgentFactCheck as a reference implementation for the LangChain pattern is excellent, and the accompanying documentation in agent_development_guide.md is very thorough and helpful. The refactoring of BaseAgent to support custom input formatting is a solid improvement. I've left a few minor comments regarding inconsistencies in configuration values and documentation details. Overall, this is a high-quality contribution that significantly improves the agent development framework.