> ## Documentation Index
> Fetch the complete documentation index at: https://docs.galtea.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Get endpoint connections

> Get list of endpoint connections with pagination and filtering. See [Endpoint Connections](https://docs.galtea.ai/concepts/product/endpoint-connection).

> [!WARNING]
>
> You are reading the docs of Galtea 5.15. [Read this page in the latest version](https://docs.galtea.ai/api-reference/endpoint-connections/get-endpoint-connections.md).


## OpenAPI

````yaml /v5.15/openapi.json get /endpointConnections
openapi: 3.0.0
info:
  version: 1.0.0
  title: Product Management Service API
  description: API documentation for Product Management Service
  contact:
    name: Galtea AI
servers:
  - url: https://api.galtea.ai
security:
  - bearerAuth: []
tags: []
externalDocs:
  description: Galtea Platform Documentation
  url: https://docs.galtea.ai
paths:
  /endpointConnections:
    get:
      tags:
        - endpoint-connections
      summary: Get endpoint connections
      description: >-
        Get list of endpoint connections with pagination and filtering. See
        [Endpoint
        Connections](https://docs.galtea.ai/concepts/product/endpoint-connection).
      operationId: getEndpointConnections
      parameters:
        - name: ids
          in: query
          style: form
          explode: true
          description: Filter by endpoint connection IDs
          required: false
          schema:
            type: array
            items:
              type: string
        - name: productIds
          in: query
          style: form
          explode: true
          description: Filter by product IDs
          required: false
          schema:
            type: array
            items:
              type: string
        - name: names
          in: query
          style: form
          explode: true
          description: Filter by endpoint connection names (exact match, multiple)
          required: false
          schema:
            type: array
            items:
              type: string
        - name: name
          in: query
          description: Filter by endpoint connection name (partial match)
          required: false
          schema:
            type: string
        - name: url
          in: query
          description: Filter by URL (partial match)
          required: false
          schema:
            type: string
        - name: types
          in: query
          style: form
          explode: true
          description: Filter by endpoint connection types
          required: false
          schema:
            type: array
            items:
              type: string
        - name: httpMethods
          in: query
          style: form
          explode: true
          description: Filter by HTTP methods
          required: false
          schema:
            type: array
            items:
              type: string
              enum:
                - GET
                - POST
                - PUT
                - PATCH
                - DELETE
        - name: authTypes
          in: query
          style: form
          explode: true
          description: Filter by authentication types
          required: false
          schema:
            type: array
            items:
              type: string
              enum:
                - NONE
                - BEARER
                - API_KEY
                - BASIC
        - name: sort
          in: query
          style: form
          explode: true
          description: Sort instructions (field and direction pairs)
          required: false
          schema:
            type: array
            items:
              type: string
        - name: limit
          in: query
          description: Maximum number of results
          required: false
          schema:
            type: integer
        - name: offset
          in: query
          description: Number of results to skip
          required: false
          schema:
            type: integer
        - name: fromCreatedAt
          in: query
          description: >-
            Filter endpoint connections created at or after this timestamp (ISO
            8601 format)
          required: false
          schema:
            type: string
            format: date-time
        - name: toCreatedAt
          in: query
          description: >-
            Filter endpoint connections created at or before this timestamp (ISO
            8601 format)
          required: false
          schema:
            type: string
            format: date-time
      responses:
        '200':
          description: Endpoint connections retrieved successfully
          content:
            application/json:
              schema:
                type: array
                items:
                  $ref: '#/components/schemas/EndpointConnection'
        '401':
          description: Unauthorized
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
      security:
        - bearerAuth: []
components:
  schemas:
    EndpointConnection:
      type: object
      required:
        - productId
        - name
        - url
        - httpMethod
        - authType
        - inputTemplate
        - outputMapping
      properties:
        id:
          type: string
          example: ec_123
        productId:
          type: string
          example: prod_123
        userId:
          type: string
          nullable: true
          example: user_123
        name:
          type: string
          example: My API Connection
        url:
          type: string
          example: https://api.example.com/v1/chat/completions
        httpMethod:
          type: string
          enum:
            - GET
            - POST
            - PUT
            - PATCH
            - DELETE
          example: POST
        authType:
          type: string
          enum:
            - NONE
            - BEARER
            - API_KEY
            - BASIC
          example: BEARER
        hasAuthToken:
          type: boolean
          description: Whether an auth token is configured (token value is not exposed)
        username:
          type: string
          nullable: true
          description: Username for Basic auth
        hasPassword:
          type: boolean
          description: Whether a password is configured (password value is not exposed)
        headers:
          type: object
          nullable: true
          additionalProperties:
            type: string
          description: >-
            Custom headers as key-value pairs. A secret value is encrypted at
            rest once its headers are saved, and every read returns it as
            "••••••••". A value is secret unless its name is Content-Type,
            Accept, Accept-Language, Accept-Encoding, Cache-Control, User-Agent,
            traceparent or tracestate, or it holds only placeholders (for
            example "Bearer {{ bearer_token }}"). On update, send "••••••••" to
            keep the stored value of that header name; any other value replaces
            it. Sending "••••••••" for a header with no stored value is a 400,
            and so is a value in the stored ciphertext shape or with a character
            a header cannot carry. A create or update that sends headers holding
            `traceparent` is a 400, because Galtea sets it on every turn call; a
            stored one is dropped on send.
          example:
            Content-Type: application/json
        inputTemplate:
          type: string
          nullable: true
          description: >-
            Jinja2/Nunjucks template for request body. Must contain {{
            input.user_message }} placeholder.
          example: '{"prompt": "{{ input.user_message }}"}'
        inputSchema:
          type: object
          nullable: true
          description: >-
            Optional JSON Schema (draft 2020-12) defining the structure of input
            fields, their types, and valid values. Must have type "object" with
            a "properties" key.
          example:
            type: object
            properties:
              user_message:
                type: string
                description: The main user message to the agent
              reasoning_level:
                type: string
                enum:
                  - low
                  - medium
                  - high
                description: How much reasoning the agent should apply
              topic:
                type: string
                enum:
                  - billing
                  - shipping
                  - returns
                  - account
                description: Conversation topic
              customer_id:
                type: string
                pattern: ^C-\d{3,6}$
                description: Customer identifier
            required:
              - user_message
        outputMapping:
          type: object
          nullable: true
          additionalProperties:
            type: string
          description: >-
            JSONPath expressions to extract values from API response. Must
            include "output" key. Special keys: "retrieval_context",
            "session_id" (initialization), "traces", and "output_files".
            "output_files" matches one file or a list of files the endpoint
            answers with, each a data: URI, an http(s) URL, the storage URI of a
            Galtea file, or an object with "base64", "url" or "uri" plus
            optional "mime_type" and "filename". Each file is stored as a Galtea
            file on the turn output: at most 20 files and 20 MB per output, an
            accepted file type (the file extension decides it), and a URL is
            downloaded once without the connection headers or credentials, over
            plain http only when the connection acknowledged it. A Galtea file
            is linked rather than copied, and must be an uploaded file of the
            product organization. A file that cannot be stored fails the turn. A
            metric scores an output file only when its judge model reads that
            kind of file (PNG, JPEG, WebP, GIF or PDF); any other metric that
            reads the output is skipped, except a human evaluation or
            self-hosted one, and the reason names the file. Every other key is
            stored on the session metadata.
          example:
            output: $.choices[0].message.content
            output_files: $.files
        type:
          type: string
          enum:
            - INITIALIZATION
            - CONVERSATION
            - FINALIZATION
          example: CONVERSATION
        timeout:
          type: integer
          nullable: true
          minimum: 1
          maximum: 120
          default: 15
          example: 15
          description: >-
            Per-attempt request timeout in seconds. Unset applies the default
            (15). Capped server-side at 120 seconds.
        rateLimit:
          type: integer
          nullable: true
          minimum: 1
          example: 60
          description: Maximum requests per minute. Unset (null) means no rate limit.
        retryEnabled:
          type: boolean
          default: true
          description: Whether automatic retry is enabled for failed requests
        retryMaxAttempts:
          type: integer
          nullable: true
          minimum: 0
          maximum: 10
          default: 1
          description: >-
            Maximum number of retry attempts (applied when retryEnabled is
            true). 0 is only valid when retryEnabled is false; with retry
            enabled the minimum is 1.
        retryInitialDelayMs:
          type: integer
          nullable: true
          minimum: 0
          default: 1000
          description: Initial delay in milliseconds before first retry
        retryBackoffStrategy:
          type: string
          nullable: true
          enum:
            - FIXED
            - EXPONENTIAL
            - LINEAR
          default: EXPONENTIAL
          description: Strategy for increasing delay between retries
        retryMaxDelayMs:
          type: integer
          nullable: true
          minimum: 0
          maximum: 60000
          default: 3000
          description: >-
            Maximum delay between retry attempts, in milliseconds. Capped
            server-side at 60000 (60 seconds) to bound retry blast radius.
        retryableStatusCodes:
          type: array
          items:
            type: integer
            minimum: 100
            maximum: 599
          default:
            - 429
            - 500
            - 502
            - 503
            - 504
          description: HTTP status codes that should trigger a retry
        acknowledgedInsecureHttp:
          type: boolean
          default: false
          description: >-
            Required true to use an http:// URL. It records that the creator
            accepted that traffic to this endpoint is not encrypted, so anyone
            on the network path can read prompts, responses and credentials. The
            acceptance covers the connection, not one host: a redirect may send
            an acknowledged request to any public http:// host, carrying the
            connection credentials with it. Without it every hop the platform
            makes to the endpoint must stay on https. It is stored as false for
            an https:// URL, whatever the request sends.
        conversationEndRules:
          type: array
          maxItems: 20
          default: []
          description: >-
            Responses that mean the agent closed the conversation on its own.
            Only a CONVERSATION connection accepts rules. A rule matches a
            response with its status code and, when bodyMatch is set, a response
            body whose first JSONPath match equals the expected value as text.
            The rules are checked before the retry policy, and a matching
            response is never retried, even when its status code is in
            retryableStatusCodes. A matching 2xx response with an agent output
            keeps its turn as the last one, on any turn and in a single-turn
            test. A matching response with no agent output (a non-2xx response,
            or a 2xx response with no output) removes the rejected turn on a
            later turn of a scenario test. In both cases the session completes
            with the turns so far and is evaluated, with the stopping reason
            "The agent ended the conversation (HTTP <code>)". A match with no
            agent output on the first turn, or on a single-turn test, still
            fails the session. An empty list keeps the default behaviour.
          items:
            type: object
            required:
              - statusCode
            properties:
              statusCode:
                type: integer
                minimum: 100
                maximum: 599
                description: >-
                  HTTP status code of the response. A 2xx status code needs a
                  bodyMatch.
                example: 409
              bodyMatch:
                type: object
                nullable: true
                required:
                  - path
                  - equals
                description: >-
                  Optional check on the response body. A body that is not JSON,
                  or has no value at the path, does not match.
                properties:
                  path:
                    type: string
                    description: >-
                      JSONPath into the response body, starting with $.
                      Surrounding spaces are trimmed, and filter expressions are
                      refused. Only the first match is compared, and only a
                      string, number or boolean value.
                    example: $.error.code
                  equals:
                    type: string
                    description: >-
                      Expected value, compared as text (a number 409 matches
                      "409").
                    example: conversation_ended
        createdAt:
          type: string
          format: date-time
        deletedAt:
          type: string
          format: date-time
          nullable: true
    Error:
      type: object
      properties:
        error:
          type: string
          example: Error type
        message:
          type: string
          example: Error message description
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: >-
        API key authorization. Pass your API key in the Authorization header as
        a Bearer token. Both new (`gsk_*`) and legacy (`gsk-`) API keys are
        accepted, e.g. `Authorization: Bearer gsk_...` or `Authorization: Bearer
        gsk-...`.

````

This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.