fix: ignore disconnected agentflow nodes

[iokinpardo]

Agentflow disconnected node tolerance

• Purpose / What it does: Allows Agentflow executions to ignore nodes on the canvas that are not connected to the Start node, so unfinished ideas can remain while testing.
• Usage example: Leave experimental Agent or Condition nodes detached from Start and run the flow—only the connected graph executes.
• Dependencies / breaking changes: None.

[gemini-code-assist]

Summary of Changes

Hello @iokinpardo, 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 flexibility of Agentflow by introducing tolerance for disconnected nodes. It allows developers to maintain experimental or incomplete parts of their workflow on the canvas without them interfering with the execution of the primary, connected flow. This change improves the development and testing experience by making the Agentflow more forgiving and adaptable to iterative design processes.

Highlights

• Agentflow Execution Logic: Implemented a new feature that allows Agentflow executions to ignore nodes on the canvas that are not connected to the 'Start' node. This enables users to keep experimental or unfinished nodes on the canvas without affecting the execution of the main flow.
• Graph Traversal Utility: Introduced getReachableNodesFromStart and filterDisconnectedStartingNodes functions in buildAgentflow.ts to identify and filter out nodes that are not part of the main connected graph starting from the startAgentflow node.
• Documentation Update: The README.md has been updated to include a new 'Features' section, detailing the purpose and usage example of the 'Agentflow disconnected node tolerance' feature.
• New Test Coverage: Added a new test file (build-agentflow.util.test.ts) to specifically validate the functionality of filtering disconnected starting nodes, ensuring the new logic works as expected.

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 introduces a useful feature to ignore nodes that are not connected to the startAgentflow node, allowing for a cleaner development experience. The implementation correctly identifies reachable nodes using a graph traversal. However, I've found a potential issue in the fallback logic that could cause disconnected nodes to execute in certain edge cases, which contradicts the feature's purpose. I've also pointed out a small piece of redundant code that can be removed for simplification. Additionally, a new test file has been added to validate the new functionality, which is great.

[gemini-code-assist]

The fallback logic in this return statement appears to contradict the feature's goal. If filteredStartingNodes is empty (meaning no 0-in-degree nodes are connected to the startAgentflow node), the function currently returns the original startingNodeIds, which includes disconnected nodes. This would cause the disconnected nodes to execute, which is the opposite of what's intended.

The function should simply return filteredStartingNodes. If the array is empty, it correctly signals that there are no valid starting points for the execution, and the flow will not run.

    return filteredStartingNodes

[gemini-code-assist]

This check for !reachableNodes.size is redundant. The getReachableNodesFromStart function is guaranteed to return a Set containing at least the startNodeId when a valid ID is provided. Since you've already confirmed startAgentflowNode exists before this call, reachableNodes will never be empty. This block can be safely removed to simplify the code.