Feature Description
Feature: Execution Debug & Trace Panel
Problem
When working with complex flows, especially multi-node LLM pipelines, it is difficult to understand the following:
- What happens at each node during execution
- Input/output transformations
- Where exactly failures occur
- Performance bottlenecks
Currently, debugging requires trial-and-error, which slows down development.
Proposed Solution
Introduce a Debug & Trace Panel for flow execution:
Core Features
- Step-by-step execution trace of nodes
- For each node:
- Input payload
- Output result
- Execution duration
- Highlight failed nodes with error details
- Visual execution order (timeline or sequence)
Optional Controls
- Toggle debug mode ON/OFF
- Minimal overhead when disabled
Implementation Idea
- Hook into node execution pipeline
- Capture metadata (input/output/time/error)
- Store temporarily during execution
- Render in a side panel (similar to dev tools)
Benefits
- Massive improvement in developer experience
- Faster debugging of AI workflows
- Better visibility for production-level flows
- Especially useful for complex agent pipelines
Additional Context
This would position Flowise closer to production-grade observability seen in modern workflow/AI orchestration tools.
I’d like to work on this feature and can start with a minimal version (basic trace + node I/O logging) and iterate further.
Let me know your thoughts or any preferred direction.
Feature Category
Workflow/Flow Management
Problem Statement
Currently, debugging flow execution in Flowise is difficult, especially for multi-node LLM pipelines.
There is no clear visibility into:
- Node-by-node execution flow
- Input and output data transformations
- Exact failure points
- Performance bottlenecks
This results in trial-and-error debugging, making development slower and less efficient, particularly for complex workflows.
Proposed Solution
Introduce a Debug & Trace Panel that provides visibility into flow execution.
Key features:
- Step-by-step execution trace of nodes
- Display input/output data for each node
- Show execution duration per node
- Highlight failed nodes with error details
- Visual execution sequence (timeline or ordered list)
Optional:
- Toggle debug mode ON/OFF
- Minimal performance overhead when disabled
Implementation approach:
- Hook into node execution lifecycle
- Capture execution metadata (input/output/time/error)
- Store temporarily during execution
- Render in a side panel in the UI
Mockups or References
Similar concepts can be seen in:
- Node-RED debug panel
- Browser DevTools (network/request tracing)
- Workflow tools with execution logs
A simple initial version can start with a basic panel showing node logs and execution order and later expand into a full visual timeline.
Additional Context
This feature can be implemented incrementally:
Phase 1:
- Basic execution logging (node input/output + order)
Phase 2:
- UI panel for visualization
Phase 3:
- Advanced tracing (timeline, performance insights)
Happy to start with Phase 1 and iterate based on feedback.
Feature Description
Feature: Execution Debug & Trace Panel
Problem
When working with complex flows, especially multi-node LLM pipelines, it is difficult to understand the following:
Currently, debugging requires trial-and-error, which slows down development.
Proposed Solution
Introduce a Debug & Trace Panel for flow execution:
Core Features
Optional Controls
Implementation Idea
Benefits
Additional Context
This would position Flowise closer to production-grade observability seen in modern workflow/AI orchestration tools.
I’d like to work on this feature and can start with a minimal version (basic trace + node I/O logging) and iterate further.
Let me know your thoughts or any preferred direction.
Feature Category
Workflow/Flow Management
Problem Statement
Currently, debugging flow execution in Flowise is difficult, especially for multi-node LLM pipelines.
There is no clear visibility into:
This results in trial-and-error debugging, making development slower and less efficient, particularly for complex workflows.
Proposed Solution
Introduce a Debug & Trace Panel that provides visibility into flow execution.
Key features:
Optional:
Implementation approach:
Mockups or References
Similar concepts can be seen in:
A simple initial version can start with a basic panel showing node logs and execution order and later expand into a full visual timeline.
Additional Context
This feature can be implemented incrementally:
Phase 1:
Phase 2:
Phase 3:
Happy to start with Phase 1 and iterate based on feedback.