Visual Agent Logic Builder for Conditional Branching (If/Then Logic)

I'd like to request a feature similar to what Relevance.ai calls a "Flow Builder" (docs: https://relevanceai.com/docs/agent/build-your-agent/agent-settings/flow-builder) for designing agent behavior within AgenticFlow.

Currently, much of an agent's decision-making and task execution path is guided by its system prompt and the inherent capabilities of its LLM when choosing tools (MCPs or Workflows). A visual way to define conditional logic within the agent's own operational flow would be incredibly powerful and have a more granular control of the agent’s action.

Proposed Functionality ("Agent Logic Builder" / "Decision Tree Builder"):

1. Define Conditions: Set up "IF" statements based on various triggers or data points, for example:

   • User input (e.g., keywords, sentiment, intent detected).

   • Output from a previous agent step or tool (e.g., data retrieved from an API call, content of a scraped page).

   • Presence or value of specific variables.

   • Chat history context.

2. Create Branches: Based on whether a condition is TRUE or FALSE, define different paths for the agent to follow ("THEN" / "ELSE").

3. Assign Actions/Instructions per Branch: For each branch, specify:

   • Which tool (MCP action or Workflow) the agent should use next.

   • Specific sub-prompts or instructions to provide to an LLM node or the agent's next thought process.

   • Variables to set or modify.

   • Direct responses for the agent to give.

4. Nested Logic: Allow for nesting conditions to create complex, multi-step decision trees.