How to Create and Edit Pipelines from Canvas¶

Introduction¶

This user guide provides a comprehensive overview of the Pipeline Canvas feature in ELITEA and how to create, configure, and manage pipelines directly from the chat interface. The Pipeline Canvas enables streamlined pipeline management with an intuitive interface, enhancing productivity and collaboration across different AI workflow scenarios.

Pipeline Canvas Feature: Visual Pipeline Management¶

The Pipeline Canvas interface in ELITEA serves as an integrated pipeline management system accessible directly from the chat interface. This feature enables you to create, configure, and manage AI pipelines without leaving your conversation context, streamlining your development workflow.

Key Aspects of the Pipeline Canvas Feature:

Integrated Chat Experience: Access pipeline management directly from the PARTICIPANTS section in chat, maintaining conversation context while managing AI workflows.
Dual-Mode Configuration: Use both Configuration mode for general settings and Flow Editor mode for visual workflow design.
Real-time Validation: Configuration fields are validated in real-time, ensuring proper setup before pipeline creation.
Instant Integration: Created pipelines are immediately available for use in conversations and can be added to the PARTICIPANTS section.
Advanced Model Configuration: Integration with ELITEA's LLM model management system for comprehensive AI model selection and fine-tuning.

Creating Pipelines via Canvas Interface¶

Step 1: Access the Pipeline Creation Canvas¶

Navigate to the Chat page (main sidebar menu).
In the PARTICIPANTS section, locate Pipelines.
Click on the Create new pipeline button.

The "Create New Pipeline" canvas interface will be displayed with all available configuration sections.

Step 2: Configure General Information¶

The pipeline creation interface is organized into two main tabs: Configuration and Flow editor. Start with the Configuration tab:

Required Fields:

Name* (required): Enter a unique, descriptive name for your pipeline (e.g., "Content Review Workflow", "Data Processing Pipeline")
Description* (required): Provide a clear description of what your pipeline will do and its purpose
Tags (optional): Add relevant tags by typing tag names to help organize and categorize your pipelines

Step 3:: Configure Welcome Message (Optional)¶

In the WELCOME MESSAGE section, add a message that users see when they first interact with your pipeline:

Example: "Hello! I'm your content review pipeline. I'll help you process and validate your content through multiple review stages."

Step 4: Set Up Conversation Starters (Optional)¶

In the CONVERSATION STARTERS section:

Click the + Starter button to add conversation starters (maximum 4 starters allowed)
Enter helpful prompts that guide users on how to interact with your pipeline
Each starter has a delete (trash) icon for removal if needed
Examples:
"Start a new review process"
"Check pipeline status"

Step 5: Configure Advanced Settings (Optional)¶

In the ADVANCED section:

Step limit: Set the maximum number of steps the pipeline can execute (default: 25)
This controls how many operations the pipeline can perform in a single execution
Allow Attachments: Toggle to enable or disable file attachment support for the pipeline
When enabled, users can attach files and documents when interacting with the pipeline in conversations
Requires configuring an artifact toolkit to store attached files
Click the settings icon (⚙️) next to the toggle to:
- Select an existing artifact toolkit from your project
- Create a new artifact toolkit for file storage
Useful for pipelines that need to process documents, images, or other file types as input

Note

Required fields are marked with an asterisk * and must be completed before the pipeline can be created.

Step 6: Save Initial Configuration¶

Click the Save button to create your pipeline with the basic configuration. This will proceed to the advanced configuration interface where you can:

Configure LLM model and settings
Add toolkits, agents, pipelines, and MCP connections
Access the Flow Editor for visual workflow design

Advanced Pipeline Configuration¶

After saving the initial configuration, the advanced configuration interface opens with two main tabs:

Configuration Tab¶

LLM Model and Settings Configuration¶

Model Selection:
- Click the Select LLM Model button at the top of the interface
- Choose from available LLM models in your project (e.g., "gpt-4o", "gpt-4o-mini")
- The selected model will be displayed on the button

Model Settings:
- Click the Model Settings button (gear icon) next to the model selector
- Configure the following parameters:
- Temperature (0.1-1.0): Controls creativity and randomness in responses
  - Lower values (0.1-0.3): More focused and deterministic responses
  - Higher values (0.7-1.0): More creative and varied responses
- Top P (0.1-1.0): Controls diversity of word choices
- Max Completion Tokens: Maximum length of the response (e.g., 4096)
- Click Apply to save the model settings

Toolkits Configuration¶

In the TOOLKITS section, you can enhance your pipeline's capabilities by adding various integrations:

Adding Toolkits:
- Click + Toolkit to select from available toolkits or create new ones
- Browse and select toolkits like GitHub, Jira, Confluence, etc.

Adding MCP Connections:
- Click + MCP to add Model Context Protocol connections
- Select from available MCP configurations
- MCP connections appear in the toolkits list with proper status indicators

Adding Nested Agents:
- Click + Agent to add agents as pipeline components
- Select from existing agents in your project
- Choose specific versions from the version dropdown
- Nested agents provide AI capabilities within pipeline nodes

Adding Nested Pipelines:
- Click + Pipeline to integrate other pipeline workflows
- Select from available pipelines in your project
- Nested pipelines enable complex workflow compositions
Allow Attachments Toggle:
- Enable or disable file attachment support for the pipeline

Flow Editor Tab¶

The Flow Editor tab provides a visual interface for designing your pipeline workflow. This tab is only available after the initial pipeline save.

Flow Editor Availability

The Flow Editor is disabled during pipeline creation. You must save the pipeline first in the Configuration tab before accessing the Flow Editor.

Flow Editor Features:¶

Visual Node Management:
- Drag and drop nodes onto the canvas
- Connect nodes to create workflow paths
- Configure individual node properties
- Click the + (Add Node) button to add available nodes to your workflow
- Multiple node types available for different workflow operations (Agent, Decision, Loop, Router, State Modifier, and more)
State Management:
- State: The pipeline's state holds data that flows through your workflow
- State Variables: Define and manage variables that store information between nodes
- State Modifier Node: Update, transform, or clean up state variables during workflow execution
- Use state variables to pass data between different nodes in your pipeline
- Configure input and output variables for each node to control data flow
YAML Synchronization:
- Changes in the Flow Editor automatically update the YAML instructions
- Manual YAML edits can also be reflected in the visual editor
- Seamless switching between visual and code-based editing
Editor Modes:
- Flow mode: Visual drag-and-drop interface
- YAML mode: Direct YAML code editing

Step 7: Finalize Pipeline Creation¶

Once you have completed configuring your pipeline:

Click the Save button to save all your configuration changes
Click the X button to close the canvas interface

Your newly created pipeline will appear in the PARTICIPANTS section under Pipelines and becomes immediately available for use in conversations.

Editing Pipelines via Canvas Interface¶

Accessing Pipeline Edit Mode¶

There are two ways to access the pipeline edit mode:

From PARTICIPANTS Section
- Navigate to the Chat page where the pipeline is available.
- In the PARTICIPANTS section, locate Pipelines.
- Find the pipeline you want to edit.
- Hover over the pipeline to reveal action buttons.
- Click the pencil Edit icon that appears.
From Chat Interface
- When a pipeline is active in your conversation, click the Pipeline Settings button (gear icon) that appears in the chat interface.
- This will open the pipeline configuration canvas directly.

The pipeline configuration canvas will open with current settings pre-populated.

Modifying Pipeline Configuration¶

Once in edit mode, you can modify any configuration parameters to update your pipeline settings. The edit interface displays current settings pre-populated, allowing you to make changes to:

Configuration Tab Modifications:¶

Welcome Message: Change the initial message users see
Conversation Starters: Add, remove, or modify starter prompts (maximum 4)
Advanced: Adjust step limits and other parameters
LLM Model and Settings: Switch models or fine-tune model parameters (temperature, top-p, max tokens)

Toolkits and Integrations:

Add or remove toolkits, MCP connections, nested agents, and nested pipelines
Nested Toolkits: View and manage individual tools within each toolkit
MCP Servers: View and manage available tools for each MCP connection
Nested Agents: Select and change specific versions from the version dropdown
Nested Pipelines: Select and change specific versions from the version dropdown

Flow Editor Tab Modifications:¶

Visual Workflow Design: Add, remove, or rearrange nodes in the flow
Node Configuration: Update properties for individual nodes
Connection Management: Modify transitions between nodes
YAML Editing: Direct editing of the workflow YAML code

For example, if you have a pipeline configured for content processing, you can easily update the Instructions to include new processing stages, add additional Toolkits like GitLab for repository integration, or redesign the workflow in the Flow Editor to add parallel processing paths.

Changes are reflected immediately in the interface. After making your desired configuration changes, use the Save button located in the top right corner to apply your modifications. The updated configuration becomes immediately applied.

Version Management¶

When editing pipelines from the chat interface:

Version Selector:
- The current version is displayed on the version dropdown in the pipeline settings panel
- The current version is also displayed on the pipeline card in the PARTICIPANTS sidebar
- Click the version dropdown to view all available versions of the pipeline
- Select different versions to view or edit
Version Status:
- Draft versions: Can be edited and modified
- Published versions: Read-only, cannot be edited

Published Version Limitations

Published versions are read-only and cannot be edited. To modify a published version, create a new version or edit an existing draft version.

Troubleshooting Common Issues¶

Pipeline Creation Issues¶

Missing Required Fields:

Problem: Save button remains disabled
Solution: Review all fields marked with asterisk (*) and ensure they are completed
Verification: Check that both Name and Description fields contain valid text

YAML Syntax Errors:

Problem: Pipeline creation fails due to invalid YAML
Solution: Verify YAML syntax in the Instructions field, ensure proper indentation and structure
Prevention: Use YAML validators or start with a simple workflow template

Model Configuration Errors:

Problem: Pipeline creation fails due to model configuration issues
Solution: Verify that a valid LLM model is selected and model settings are within acceptable ranges
Prevention: Use default model settings initially, then customize as needed

Pipeline Management Issues¶

Edit Mode Access:

Problem: Cannot access edit mode for existing pipeline
Solution: Verify you have appropriate permissions (applications.update) and the pipeline version is not published
Workaround: Ensure no active operations are using the pipeline

Flow Editor Not Available:

Problem: Flow Editor tab is disabled
Solution: Save the pipeline first in the Configuration tab, then the Flow Editor will become available
Resolution: Flow Editor is only accessible after initial pipeline creation

Configuration Validation:

Problem: Changes cannot be saved due to validation errors
Solution: Review error messages and ensure all required fields maintain valid values
Resolution: Check that external resources (models, toolkits, agents) are still accessible

PARTICIPANTS Section Issues¶

Pipeline Not Appearing:

Problem: Created pipeline doesn't appear in PARTICIPANTS
Solution: Refresh the interface or verify the pipeline was created in the correct workspace/project
Check: Ensure you're viewing the correct project context

Model Settings Not Applying:

Problem: Model configuration changes don't take effect
Solution: Ensure you clicked "Apply" in the model settings dialog before closing
Prevention: Always use the Apply button rather than just closing the settings dialog

Version Switching Issues:

Problem: Cannot switch between pipeline versions
Solution: If you have unsaved changes, save or discard them before switching versions
Note: Switching versions with unsaved changes will prompt a confirmation dialog

Conclusion¶

The Pipeline Canvas feature in ELITEA provides a powerful and intuitive way to manage AI pipeline creation and configuration directly from your chat interface. By following the practices outlined in this guide, you can effectively create, configure, and manage pipelines that enhance your development workflows and team collaboration.

Key Benefits Recap:

Streamlined Creation: Create pipelines without leaving your conversation context
Dual-Mode Configuration: Use both Configuration and Flow Editor for flexible design
Visual Workflow Design: Leverage the Flow Editor for intuitive pipeline creation
Real-time Integration: Immediately use pipelines in your conversations
Advanced AI Configuration: Leverage comprehensive LLM model management and fine-tuning
Team Collaboration: Share and manage pipelines within team projects