Module: OpenAI::Models::Responses::Tool::Mcp::AllowedCaller

Extended by:
Internal::Type::Enum
Defined in:
lib/openai/models/responses/tool.rb,
sig/openai/models/responses/tool.rbs

Constant Summary collapse

DIRECT =

Returns:

  • (:direct) 
:direct
PROGRAMMATIC =

Returns:

  • (:programmatic) 
:programmatic

Instance Method Summary collapse

Methods included from Internal::Type::Enum

==, ===, coerce, dump, hash, inspect, to_sorbet_type, values, values

Methods included from Internal::Util::SorbetRuntimeSupport

#const_missing, #define_sorbet_constant!, #sorbet_constant_defined?, #to_sorbet_type, to_sorbet_type

Methods included from Internal::Type::Converter

#coerce, coerce, #dump, dump, #inspect, inspect, meta_info, new_coerce_state, type_info

Instance Method Details

#initialize(server_label:, allowed_callers: nil, allowed_tools: nil, authorization: nil, connector_id: nil, defer_loading: nil, headers: nil, require_approval: nil, server_description: nil, server_url: nil, tunnel_id: nil, type: :mcp) ⇒ Object

Some parameter documentations has been truncated, see OpenAI::Models::Responses::Tool::Mcp for more details.

Give the model access to additional tools via remote Model Context Protocol (MCP) servers. Learn more about MCP.

Parameters:

  • server_label (String) —

    A label for this MCP server, used to identify it in tool calls.

  • allowed_callers (Array<Symbol, OpenAI::Models::Responses::Tool::Mcp::AllowedCaller>, nil) (defaults to: nil) —

    The tool invocation context(s).

  • allowed_tools (Array<String>, OpenAI::Models::Responses::Tool::Mcp::AllowedTools::McpToolFilter, nil) (defaults to: nil) —

    List of allowed tool names or a filter object.

  • authorization (String) (defaults to: nil) —

    An OAuth access token that can be used with a remote MCP server, either

  • connector_id (Symbol, OpenAI::Models::Responses::Tool::Mcp::ConnectorID) (defaults to: nil) —

    Identifier for service connectors, like those available in ChatGPT. One of

  • defer_loading (Boolean) (defaults to: nil) —

    Whether this MCP tool is deferred and discovered via tool search.

  • headers (Hash{Symbol=>String}, nil) (defaults to: nil) —

    Optional HTTP headers to send to the MCP server. Use for authentication

  • require_approval (OpenAI::Models::Responses::Tool::Mcp::RequireApproval::McpToolApprovalFilter, Symbol, OpenAI::Models::Responses::Tool::Mcp::RequireApproval::McpToolApprovalSetting, nil) (defaults to: nil) —

    Specify which of the MCP server's tools require approval.

  • server_description (String) (defaults to: nil) —

    Optional description of the MCP server, used to provide more context.

  • server_url (String) (defaults to: nil) —

    The URL for the MCP server. One of server_url, connector_id, or

  • tunnel_id (String) (defaults to: nil) —

    The Secure MCP Tunnel ID to use instead of a direct server URL. One of

  • type (Symbol, :mcp) (defaults to: :mcp) —

    The type of the MCP tool. Always mcp.



189
190
191
192
193
194
195
196
197
 
# File 'lib/openai/models/responses/tool.rb', line 189

module AllowedCaller
  extend OpenAI::Internal::Type::Enum

  DIRECT = :direct
  PROGRAMMATIC = :programmatic

  # @!method self.values
  #   @return [Array<Symbol>]
end

#self?.values ⇒ ::Array[OpenAI::Models::Responses::Tool::Mcp::allowed_caller]

Returns:

  • (::Array[OpenAI::Models::Responses::Tool::Mcp::allowed_caller]) 


118
 
# File 'sig/openai/models/responses/tool.rbs', line 118

def self?.values: -> ::Array[OpenAI::Models::Responses::Tool::Mcp::allowed_caller]