---
title: "Get definition of a named custom integration"
method: GET
path: "/integrations/custom/{name}/definition"
tags: ["integrations"]
---

# Get definition of a named custom integration

`GET /integrations/custom/{name}/definition`

Return the full JSON definition of a named custom integration, including decrypted secrets.

## Path parameters

- `name` string, required — Slug identifying this named custom integration

## Response `200`

Successful Response

- CustomIntegrationDefinition — Response schema for the full JSON definition of a custom integration. Returns the exact same structure used to create the integration, including decrypted sensitive fields (api_key_value, token, headers). Only accessible to users with edit permission (creator + admins).
  - `authentication_type` 'api_key' | 'none' | 'oauth2', required — Authentication types for custom integrations. Values: - none: No authentication required - oauth2: OAuth2 token-based authentication - api_key: API key header-based authentication
  - `endpoint` string, required
  - `default_model` string, nullable
  - `model_properties` PromptgalileoSchemasConfigCustomModelProperties[], nullable
    - `name` string, required — The model name used when calling the API.
    - `alias` string, nullable — The display name/alias for the model. Defaults to name.
    - `based_on` string, nullable — Alias of a built-in model whose parameter map should be used. For example, 'gpt-5.4'. Mutually exclusive with supported_parameters.
    - `supported_parameters` string[], nullable — Explicit list of parameter names this model supports (e.g., ['max_tokens', 'temperature', 'verbosity']). Each name must be a valid RunParamsMap field. Mutually exclusive with based_on.
  - `token` string, nullable
  - `api_key_header` string, nullable
  - `api_key_value` string, nullable
  - `authentication_scope` string, nullable
  - `oauth2_token_url` string, nullable
  - `headers` object, nullable
  - `custom_llm_config` CustomLLMConfig — Configuration for a custom LiteLLM handler class. Allows users to specify a custom implementation of litellm.CustomLLM that handles acompletion() calls with custom request/response transformation.
    - `file_name` string, required — Python file name containing the CustomLLM class (e.g., 'my_handler.py')
    - `class_name` string, required — Class name within the module (must be a litellm.CustomLLM subclass)
    - `init_kwargs` object, nullable — Optional keyword arguments to pass to the CustomLLM constructor
  - `custom_header_mapping` object, nullable
  - `multi_modal_config` MultiModalModelIntegrationConfig — Configuration for multi-modal capabilities (file uploads).
    - `max_files` integer, nullable — Maximum number of files allowed per request. None means no limit.
    - `max_file_size_bytes` integer, nullable — Maximum file size in bytes per file. None means no limit.

## Other responses

- `422` — Validation Error

---

[API](https://skmtc.net/galileo/apis/galileo-api-server.md) · [All operations](https://skmtc.net/galileo/apis/galileo-api-server/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/galileo/galileo-api-server/versions/933e8c8e6366/schema)