OpenAI Analytics icon

OpenAI Analytics

Use OpenAI Analytics API

Actions20

Overview

The "Create and Run Thread" operation in this OpenAI Analytics node allows users to create a new conversational thread, add one or multiple messages to it, optionally attach files, run the thread with a selected assistant, and wait for the completion of the run to obtain results. This operation is useful for automating interactions with AI assistants where you want to start a fresh conversation, provide initial context or multiple messages, and immediately get the assistant's response.

Common scenarios include:

  • Starting a new chat session with an AI assistant with predefined messages.
  • Running a thread that includes file attachments (e.g., documents or data) to provide additional context.
  • Automating workflows that require synchronous execution, waiting for the assistant's output before proceeding.
  • Overriding default system instructions for customized assistant behavior.

Practical example:

  • A user wants to create a support ticket conversation by sending an initial message describing the issue, attaching relevant logs as files, running the thread through a specific assistant configured for support, and retrieving the assistant’s response immediately.

Properties

Name Meaning
Authentication Choose which API credentials to use: either dedicated OpenAI Analytics API credentials or existing OpenAI API credentials.
Message Input Method Select how to input messages into the thread: "Single Message" to add only one initial message, or "Multiple Messages" to add several messages with specified roles (user or system).
Initial Message The single initial message content to add to the thread (required if using Single Message input method).
Messages A list of messages to add to the thread when using Multiple Messages input method. Each message includes a role ("User" or "System") and content text (required).
File Attachment Method How to attach files to the thread: use existing uploaded file IDs, upload binary files from the previous node, use both methods, or no file attachments.
Binary Property The name of the binary property containing file data from the previous node (required if uploading files).
File Purpose Purpose of uploaded files: for use with assistants or as input to assistants.
Enable File Attachments Whether to attach existing uploaded files (boolean).
File IDs List of existing file IDs to attach to the thread (only shown if attaching existing files and enabled).
Select Assistant How to select the assistant to run the thread: choose from a list or enter an assistant ID manually.
Assistant The assistant selected from the list to use for the run (required if selecting from list).
Assistant ID The ID of the assistant to use for the run (required if entering by ID).
Wait for Completion Whether to wait for the thread run to complete before returning the result (boolean).
Maximum Poll Time (Seconds) Maximum time in seconds to wait for the run to complete if waiting for completion.
Instructions Optional override for the default system instructions of the assistant.
Simplify Output Whether to return a simplified response format instead of the full detailed output (boolean).

Output

The node outputs the result of creating and running the thread. The main output is a JSON object containing details about the thread run, including:

  • The thread ID and run ID.
  • The messages exchanged in the thread.
  • The assistant's responses.
  • Status information about the run.
  • If Simplify Output is enabled, the response is returned in a simplified format focusing on essential data.

If files are involved, the node may handle binary data uploads but does not output binary data itself; rather, it references file IDs.

Dependencies

  • Requires valid API credentials for OpenAI Analytics API or OpenAI API.
  • Needs network access to OpenAI endpoints.
  • For file uploads, expects binary data from previous nodes in the workflow.
  • Uses OpenAI SDK client initialized with the provided credentials.
  • The node relies on internal helper modules for handling threads, assistants, files, text, embeddings, and reports.

Troubleshooting

  • Credential Errors: If credentials are missing or invalid, the node will fail to authenticate. Ensure correct API keys and optional organization IDs are set up in n8n credentials.
  • File Attachment Issues: When uploading files, ensure the binary property name matches the incoming binary data. Also, verify that file IDs used exist and are accessible.
  • Assistant Selection: Selecting an assistant by ID requires a valid assistant ID. Using the list option depends on successful retrieval of available assistants.
  • Timeouts: If Wait for Completion is enabled, the node polls for run completion up to the maximum poll time. Long-running runs might exceed this timeout, causing incomplete results.
  • Message Input Validation: When using multiple messages, each message must have a role and non-empty content. Missing required fields will cause errors.
  • API Rate Limits: Frequent calls to OpenAI APIs may hit rate limits; consider adding delays or error handling for retries.

Links and References

This summary is based solely on static analysis of the provided source code and property definitions without runtime execution.

Discussion