Deployment and upgradesOfficial documentation5 min readDeployment and upgrades

Set Up Local FastGPT Plugin Remote Connections

This remote debug workflow supports FastGPT self-hosted plugin development, allowing developers to connect their local plugin environments to a live FastG…

Overview

This remote debug workflow supports FastGPT self-hosted plugin development, allowing developers to connect their local plugin environments to a live FastGPT instance. This removes the requirement to package and deploy plugins before initial testing, accelerating the development validation cycle. After a successful connection, the local FastGPT CLI tool sends plugin metadata through the FastGPT Gateway, making the local plugin visible within the FastGPT platform for hands-on testing.

Step-by-Step Connection Workflow

Follow these exact steps to establish a remote debug connection:

  1. Access the plugin debug entry within your FastGPT instance, enable the debug channel, and copy the automatically generated connection URL.
  2. Open a terminal window, navigate to the local directory containing your FastGPT plugin code, then run the following command. Replace <connectionUrl> with the full URL you copied in the first step: ``bash fastgpt-plugin dev --connect '<connectionUrl>' ``

Upon successful connection, the local CLI will complete metadata syncing via the FastGPT Gateway, making your local plugin available in the FastGPT platform.

Debug Source Format

All connected local plugins are displayed in FastGPT under a dedicated debug source with a standardized, fixed format:

debug:tmbId:{tmbId}

This debug source is tied to the FastGPT instance used to generate the initial connection URL, ensuring the local plugin is only accessible within the correct workspace context within FastGPT.

Source: FastGPT official source

Applicability and version scope

Use this page for the documented Deployment and upgrades scenario. Confirm the FastGPT, dependency, API, and deployment versions in the official source before applying a change.

Safety guardrails

Use [REDACTED_CREDENTIAL] for credentials and private data. Confirm the documented environment and version before review.

Rollback guidance

Restore the prior technical-content authority snapshot. Restore saved configuration and data snapshots, then repeat the smallest verification scenario.