mistral-agents - Pinkfish - AI Agents & Workflows for Getting Work Done

Server path: /mistral-agents | Type: Application | PCID required: Yes

Tools

Tool	Description
`mistral_agents_api_v1_agents_create`	Create a agent that can be used within a conversation.
`mistral_agents_api_v1_agents_create_or_update_alias`	Create or update an agent version alias.
`mistral_agents_api_v1_agents_delete`	Delete an agent entity.
`mistral_agents_api_v1_agents_delete_alias`	Delete an agent version alias.
`mistral_agents_api_v1_agents_get`	Retrieve an agent entity.
`mistral_agents_api_v1_agents_get_version`	Retrieve a specific version of an agent.
`mistral_agents_api_v1_agents_list`	List agent entities.
`mistral_agents_api_v1_agents_list_version_aliases`	List all aliases for an agent.
`mistral_agents_api_v1_agents_list_versions`	List all versions of an agent.
`mistral_agents_api_v1_agents_update`	Update an agent entity.
`mistral_agents_api_v1_agents_update_version`	Update an agent version.
`mistral_agents_api_v1_conversations_append`	Append new entries to an existing conversation.
`mistral_agents_api_v1_conversations_delete`	Delete a conversation.
`mistral_agents_api_v1_conversations_get`	Retrieve a conversation information.
`mistral_agents_api_v1_conversations_history`	Retrieve all entries in a conversation.
`mistral_agents_api_v1_conversations_list`	List all created conversations.
`mistral_agents_api_v1_conversations_messages`	Retrieve all messages in a conversation.
`mistral_agents_api_v1_conversations_restart`	Restart a conversation starting from a given entry.
`mistral_agents_api_v1_conversations_start`	Create a conversation and append entries to it.

mistral_agents_api_v1_agents_create

Create a agent that can be used within a conversation. Parameters:

Parameter	Type	Required	Default	Description
`completion_args`	object	No	—	White-listed arguments from the completion API
`description`	string	No	—	The description value
`guardrails`	object[]	No	—	The guardrails value
`handoffs`	string[]	No	—	The handoffs value
`instructions`	string	No	—	Instruction prompt the model will follow during the conversation.
`metadata`	object	No	—	Custom type for metadata with embedded validation.
`model`	string	Yes	—	The model value
`name`	string	Yes	—	The name value
`tools`	any[]	No	—	List of tools which are available to the model during the conversation.
`version_message`	string	No	—	Version Message

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "completion_args": {
      "type": "object",
      "description": "White-listed arguments from the completion API",
      "properties": {
        "stop": {
          "description": "The stop value"
        },
        "presence_penalty": {
          "type": "number",
          "description": "Presence Penalty"
        },
        "frequency_penalty": {
          "type": "number",
          "description": "Frequency Penalty"
        },
        "temperature": {
          "type": "number",
          "description": "The temperature value"
        },
        "top_p": {
          "type": "number",
          "description": "Top P"
        },
        "max_tokens": {
          "type": "integer",
          "description": "Max Tokens"
        },
        "random_seed": {
          "type": "integer",
          "description": "Random Seed"
        },
        "prediction": {
          "type": "object",
          "description": "Enable users to specify an expected completion, optimizing response times by leveraging known or predictable content."
        },
        "response_format": {
          "type": "object",
          "description": "Specify the format that the model must output. By default it will use `{ \"type\": \"text\" }`. Setting to `{ \"type\": \"json_object\" }` enables JSON mode, which guarantees the message the model generates is in JSON. When using JSON mode you MUST also instruct the model to produce JSON yourself with a system or a user message. Setting to `{ \"type\": \"json_schema\" }` enables JSON schema mode, which guarantees the message the model generates is in JSON and follows the schema you provide."
        },
        "tool_choice": {
          "type": "string",
          "description": "Tool Choice",
          "enum": [
            "auto",
            "none",
            "any",
            "required"
          ]
        },
        "reasoning_effort": {
          "type": "string",
          "description": "Controls the reasoning effort level for reasoning models. \"high\" enables comprehensive reasoning traces, \"none\" disables reasoning effort.",
          "enum": [
            "high",
            "none"
          ]
        }
      }
    },
    "description": {
      "type": "string",
      "description": "The description value"
    },
    "guardrails": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "block_on_error": {
            "type": "boolean",
            "description": "If true, return HTTP 403 and block request in the event of a server-side error"
          },
          "moderation_llm_v1": {
            "type": "object",
            "description": "Moderation Llm V1"
          },
          "moderation_llm_v2": {
            "type": "object",
            "description": "Moderation Llm V2"
          }
        }
      },
      "description": "The guardrails value"
    },
    "handoffs": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "The handoffs value"
    },
    "instructions": {
      "type": "string",
      "description": "Instruction prompt the model will follow during the conversation."
    },
    "metadata": {
      "type": "object",
      "description": "Custom type for metadata with embedded validation."
    },
    "model": {
      "type": "string",
      "description": "The model value"
    },
    "name": {
      "type": "string",
      "description": "The name value"
    },
    "tools": {
      "type": "array",
      "description": "List of tools which are available to the model during the conversation."
    },
    "version_message": {
      "type": "string",
      "description": "Version Message"
    }
  },
  "required": [
    "PCID",
    "model",
    "name"
  ]
}

mistral_agents_api_v1_agents_create_or_update_alias

Create or update an agent version alias. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	Yes	—	Agent Id
`alias`	string	Yes	—	The alias value
`version`	integer	Yes	—	The version value

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    },
    "alias": {
      "type": "string",
      "description": "The alias value"
    },
    "version": {
      "type": "integer",
      "description": "The version value"
    }
  },
  "required": [
    "PCID",
    "agent_id",
    "alias",
    "version"
  ]
}

mistral_agents_api_v1_agents_delete

Delete an agent entity. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	Yes	—	Agent Id

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    }
  },
  "required": [
    "PCID",
    "agent_id"
  ]
}

mistral_agents_api_v1_agents_delete_alias

Delete an agent version alias. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	Yes	—	Agent Id
`alias`	string	Yes	—	The alias value

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    },
    "alias": {
      "type": "string",
      "description": "The alias value"
    }
  },
  "required": [
    "PCID",
    "agent_id",
    "alias"
  ]
}

mistral_agents_api_v1_agents_get

Retrieve an agent entity. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	Yes	—	Agent Id
`agent_version`	object	No	—	Agent Version

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    },
    "agent_version": {
      "description": "Agent Version"
    }
  },
  "required": [
    "PCID",
    "agent_id"
  ]
}

mistral_agents_api_v1_agents_get_version

Retrieve a specific version of an agent. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	Yes	—	Agent Id
`version`	string	Yes	—	The version value

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    },
    "version": {
      "type": "string",
      "description": "The version value"
    }
  },
  "required": [
    "PCID",
    "agent_id",
    "version"
  ]
}

mistral_agents_api_v1_agents_list

List agent entities. Parameters:

Parameter	Type	Required	Default	Description
`page`	integer	No	—	Page number (0-indexed)
`page_size`	integer	No	—	Number of agents per page
`deployment_chat`	boolean	No	—	Deployment Chat
`sources`	string[]	No	—	The sources value
`name`	string	No	—	Filter by agent name
`search`	string	No	—	Search agents by name or ID
`id`	string	No	—	The id value
`metadata`	string	No	—	The metadata value

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "page": {
      "type": "integer",
      "description": "Page number (0-indexed)"
    },
    "page_size": {
      "type": "integer",
      "description": "Number of agents per page"
    },
    "deployment_chat": {
      "type": "boolean",
      "description": "Deployment Chat"
    },
    "sources": {
      "type": "array",
      "items": {
        "type": "string",
        "enum": [
          "api",
          "playground",
          "agent_builder_v1"
        ]
      },
      "description": "The sources value"
    },
    "name": {
      "type": "string",
      "description": "Filter by agent name"
    },
    "search": {
      "type": "string",
      "description": "Search agents by name or ID"
    },
    "id": {
      "type": "string",
      "description": "The id value"
    },
    "metadata": {
      "type": "string",
      "description": "The metadata value"
    }
  },
  "required": [
    "PCID"
  ]
}

mistral_agents_api_v1_agents_list_version_aliases

List all aliases for an agent. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	Yes	—	Agent Id

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    }
  },
  "required": [
    "PCID",
    "agent_id"
  ]
}

mistral_agents_api_v1_agents_list_versions

List all versions of an agent. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	Yes	—	Agent Id
`page`	integer	No	—	Page number (0-indexed)
`page_size`	integer	No	—	Number of versions per page

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    },
    "page": {
      "type": "integer",
      "description": "Page number (0-indexed)"
    },
    "page_size": {
      "type": "integer",
      "description": "Number of versions per page"
    }
  },
  "required": [
    "PCID",
    "agent_id"
  ]
}

mistral_agents_api_v1_agents_update

Update an agent entity. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	Yes	—	Agent Id
`completion_args`	object	No	—	White-listed arguments from the completion API
`deployment_chat`	boolean	No	—	Deployment Chat
`description`	string	No	—	The description value
`guardrails`	object[]	No	—	The guardrails value
`handoffs`	string[]	No	—	The handoffs value
`instructions`	string	No	—	Instruction prompt the model will follow during the conversation.
`metadata`	object	No	—	Custom type for metadata with embedded validation.
`model`	string	No	—	The model value
`name`	string	No	—	The name value
`tools`	any[]	No	—	List of tools which are available to the model during the conversation.
`version_message`	string	No	—	Version Message

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    },
    "completion_args": {
      "type": "object",
      "description": "White-listed arguments from the completion API",
      "properties": {
        "stop": {
          "description": "The stop value"
        },
        "presence_penalty": {
          "type": "number",
          "description": "Presence Penalty"
        },
        "frequency_penalty": {
          "type": "number",
          "description": "Frequency Penalty"
        },
        "temperature": {
          "type": "number",
          "description": "The temperature value"
        },
        "top_p": {
          "type": "number",
          "description": "Top P"
        },
        "max_tokens": {
          "type": "integer",
          "description": "Max Tokens"
        },
        "random_seed": {
          "type": "integer",
          "description": "Random Seed"
        },
        "prediction": {
          "type": "object",
          "description": "Enable users to specify an expected completion, optimizing response times by leveraging known or predictable content."
        },
        "response_format": {
          "type": "object",
          "description": "Specify the format that the model must output. By default it will use `{ \"type\": \"text\" }`. Setting to `{ \"type\": \"json_object\" }` enables JSON mode, which guarantees the message the model generates is in JSON. When using JSON mode you MUST also instruct the model to produce JSON yourself with a system or a user message. Setting to `{ \"type\": \"json_schema\" }` enables JSON schema mode, which guarantees the message the model generates is in JSON and follows the schema you provide."
        },
        "tool_choice": {
          "type": "string",
          "description": "Tool Choice",
          "enum": [
            "auto",
            "none",
            "any",
            "required"
          ]
        },
        "reasoning_effort": {
          "type": "string",
          "description": "Controls the reasoning effort level for reasoning models. \"high\" enables comprehensive reasoning traces, \"none\" disables reasoning effort.",
          "enum": [
            "high",
            "none"
          ]
        }
      }
    },
    "deployment_chat": {
      "type": "boolean",
      "description": "Deployment Chat"
    },
    "description": {
      "type": "string",
      "description": "The description value"
    },
    "guardrails": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "block_on_error": {
            "type": "boolean",
            "description": "If true, return HTTP 403 and block request in the event of a server-side error"
          },
          "moderation_llm_v1": {
            "type": "object",
            "description": "Moderation Llm V1"
          },
          "moderation_llm_v2": {
            "type": "object",
            "description": "Moderation Llm V2"
          }
        }
      },
      "description": "The guardrails value"
    },
    "handoffs": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "The handoffs value"
    },
    "instructions": {
      "type": "string",
      "description": "Instruction prompt the model will follow during the conversation."
    },
    "metadata": {
      "type": "object",
      "description": "Custom type for metadata with embedded validation."
    },
    "model": {
      "type": "string",
      "description": "The model value"
    },
    "name": {
      "type": "string",
      "description": "The name value"
    },
    "tools": {
      "type": "array",
      "description": "List of tools which are available to the model during the conversation."
    },
    "version_message": {
      "type": "string",
      "description": "Version Message"
    }
  },
  "required": [
    "PCID",
    "agent_id"
  ]
}

mistral_agents_api_v1_agents_update_version

Update an agent version. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	Yes	—	Agent Id
`version`	integer	Yes	—	The version value

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    },
    "version": {
      "type": "integer",
      "description": "The version value"
    }
  },
  "required": [
    "PCID",
    "agent_id",
    "version"
  ]
}

mistral_agents_api_v1_conversations_append

Append new entries to an existing conversation. Parameters:

Parameter	Type	Required	Default	Description
`conversation_id`	string	Yes	—	ID of the conversation to which we append entries.
`completion_args`	object	No	—	White-listed arguments from the completion API
`handoff_execution`	string	No	—	Handoff Execution
`inputs`	object	No	—	The inputs value
`store`	boolean	No	—	Whether to store the results into our servers or not.
`stream`	boolean	No	—	The stream value
`tool_confirmations`	object[]	No	—	Tool Confirmations

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "conversation_id": {
      "type": "string",
      "description": "ID of the conversation to which we append entries."
    },
    "completion_args": {
      "type": "object",
      "description": "White-listed arguments from the completion API",
      "properties": {
        "stop": {
          "description": "The stop value"
        },
        "presence_penalty": {
          "type": "number",
          "description": "Presence Penalty"
        },
        "frequency_penalty": {
          "type": "number",
          "description": "Frequency Penalty"
        },
        "temperature": {
          "type": "number",
          "description": "The temperature value"
        },
        "top_p": {
          "type": "number",
          "description": "Top P"
        },
        "max_tokens": {
          "type": "integer",
          "description": "Max Tokens"
        },
        "random_seed": {
          "type": "integer",
          "description": "Random Seed"
        },
        "prediction": {
          "type": "object",
          "description": "Enable users to specify an expected completion, optimizing response times by leveraging known or predictable content."
        },
        "response_format": {
          "type": "object",
          "description": "Specify the format that the model must output. By default it will use `{ \"type\": \"text\" }`. Setting to `{ \"type\": \"json_object\" }` enables JSON mode, which guarantees the message the model generates is in JSON. When using JSON mode you MUST also instruct the model to produce JSON yourself with a system or a user message. Setting to `{ \"type\": \"json_schema\" }` enables JSON schema mode, which guarantees the message the model generates is in JSON and follows the schema you provide."
        },
        "tool_choice": {
          "type": "string",
          "description": "Tool Choice",
          "enum": [
            "auto",
            "none",
            "any",
            "required"
          ]
        },
        "reasoning_effort": {
          "type": "string",
          "description": "Controls the reasoning effort level for reasoning models. \"high\" enables comprehensive reasoning traces, \"none\" disables reasoning effort.",
          "enum": [
            "high",
            "none"
          ]
        }
      }
    },
    "handoff_execution": {
      "type": "string",
      "description": "Handoff Execution",
      "enum": [
        "client",
        "server"
      ]
    },
    "inputs": {
      "description": "The inputs value"
    },
    "store": {
      "type": "boolean",
      "description": "Whether to store the results into our servers or not."
    },
    "stream": {
      "type": "boolean",
      "description": "The stream value",
      "enum": [
        false
      ]
    },
    "tool_confirmations": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "tool_call_id": {
            "type": "string",
            "description": "Tool Call Id"
          },
          "confirmation": {
            "type": "string",
            "enum": [
              "allow",
              "deny"
            ],
            "description": "The confirmation value"
          }
        },
        "required": [
          "tool_call_id",
          "confirmation"
        ]
      },
      "description": "Tool Confirmations"
    }
  },
  "required": [
    "PCID",
    "conversation_id"
  ]
}

mistral_agents_api_v1_conversations_delete

Delete a conversation. Parameters:

Parameter	Type	Required	Default	Description
`conversation_id`	string	Yes	—	ID of the conversation from which we are fetching metadata.

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "conversation_id": {
      "type": "string",
      "description": "ID of the conversation from which we are fetching metadata."
    }
  },
  "required": [
    "PCID",
    "conversation_id"
  ]
}

mistral_agents_api_v1_conversations_get

Retrieve a conversation information. Parameters:

Parameter	Type	Required	Default	Description
`conversation_id`	string	Yes	—	ID of the conversation from which we are fetching metadata.

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "conversation_id": {
      "type": "string",
      "description": "ID of the conversation from which we are fetching metadata."
    }
  },
  "required": [
    "PCID",
    "conversation_id"
  ]
}

mistral_agents_api_v1_conversations_history

Retrieve all entries in a conversation. Parameters:

Parameter	Type	Required	Default	Description
`conversation_id`	string	Yes	—	ID of the conversation from which we are fetching entries.

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "conversation_id": {
      "type": "string",
      "description": "ID of the conversation from which we are fetching entries."
    }
  },
  "required": [
    "PCID",
    "conversation_id"
  ]
}

mistral_agents_api_v1_conversations_list

List all created conversations. Parameters:

Parameter	Type	Required	Default	Description
`page`	integer	No	—	Page number for pagination
`page_size`	integer	No	—	Number of results per page
`metadata`	string	No	—	The metadata value

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "page": {
      "type": "integer",
      "description": "Page number for pagination"
    },
    "page_size": {
      "type": "integer",
      "description": "Number of results per page"
    },
    "metadata": {
      "type": "string",
      "description": "The metadata value"
    }
  },
  "required": [
    "PCID"
  ]
}

mistral_agents_api_v1_conversations_messages

Retrieve all messages in a conversation. Parameters:

Parameter	Type	Required	Default	Description
`conversation_id`	string	Yes	—	ID of the conversation from which we are fetching messages.

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "conversation_id": {
      "type": "string",
      "description": "ID of the conversation from which we are fetching messages."
    }
  },
  "required": [
    "PCID",
    "conversation_id"
  ]
}

mistral_agents_api_v1_conversations_restart

Restart a conversation starting from a given entry. Parameters:

Parameter	Type	Required	Default	Description
`conversation_id`	string	Yes	—	ID of the original conversation which is being restarted.
`agent_version`	object	No	—	Specific version of the agent to use when restarting. If not provided, uses the current version.
`completion_args`	object	No	—	White-listed arguments from the completion API
`from_entry_id`	string	Yes	—	From Entry Id
`guardrails`	object[]	No	—	The guardrails value
`handoff_execution`	string	No	—	Handoff Execution
`inputs`	object	No	—	The inputs value
`metadata`	object	No	—	Custom metadata for the conversation.
`store`	boolean	No	—	Whether to store the results into our servers or not.
`stream`	boolean	No	—	The stream value

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "conversation_id": {
      "type": "string",
      "description": "ID of the original conversation which is being restarted."
    },
    "agent_version": {
      "description": "Specific version of the agent to use when restarting. If not provided, uses the current version."
    },
    "completion_args": {
      "type": "object",
      "description": "White-listed arguments from the completion API",
      "properties": {
        "stop": {
          "description": "The stop value"
        },
        "presence_penalty": {
          "type": "number",
          "description": "Presence Penalty"
        },
        "frequency_penalty": {
          "type": "number",
          "description": "Frequency Penalty"
        },
        "temperature": {
          "type": "number",
          "description": "The temperature value"
        },
        "top_p": {
          "type": "number",
          "description": "Top P"
        },
        "max_tokens": {
          "type": "integer",
          "description": "Max Tokens"
        },
        "random_seed": {
          "type": "integer",
          "description": "Random Seed"
        },
        "prediction": {
          "type": "object",
          "description": "Enable users to specify an expected completion, optimizing response times by leveraging known or predictable content."
        },
        "response_format": {
          "type": "object",
          "description": "Specify the format that the model must output. By default it will use `{ \"type\": \"text\" }`. Setting to `{ \"type\": \"json_object\" }` enables JSON mode, which guarantees the message the model generates is in JSON. When using JSON mode you MUST also instruct the model to produce JSON yourself with a system or a user message. Setting to `{ \"type\": \"json_schema\" }` enables JSON schema mode, which guarantees the message the model generates is in JSON and follows the schema you provide."
        },
        "tool_choice": {
          "type": "string",
          "description": "Tool Choice",
          "enum": [
            "auto",
            "none",
            "any",
            "required"
          ]
        },
        "reasoning_effort": {
          "type": "string",
          "description": "Controls the reasoning effort level for reasoning models. \"high\" enables comprehensive reasoning traces, \"none\" disables reasoning effort.",
          "enum": [
            "high",
            "none"
          ]
        }
      }
    },
    "from_entry_id": {
      "type": "string",
      "description": "From Entry Id"
    },
    "guardrails": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "block_on_error": {
            "type": "boolean",
            "description": "If true, return HTTP 403 and block request in the event of a server-side error"
          },
          "moderation_llm_v1": {
            "type": "object",
            "description": "Moderation Llm V1"
          },
          "moderation_llm_v2": {
            "type": "object",
            "description": "Moderation Llm V2"
          }
        }
      },
      "description": "The guardrails value"
    },
    "handoff_execution": {
      "type": "string",
      "description": "Handoff Execution",
      "enum": [
        "client",
        "server"
      ]
    },
    "inputs": {
      "description": "The inputs value"
    },
    "metadata": {
      "type": "object",
      "description": "Custom metadata for the conversation."
    },
    "store": {
      "type": "boolean",
      "description": "Whether to store the results into our servers or not."
    },
    "stream": {
      "type": "boolean",
      "description": "The stream value",
      "enum": [
        false
      ]
    }
  },
  "required": [
    "PCID",
    "conversation_id",
    "from_entry_id"
  ]
}

mistral_agents_api_v1_conversations_start

Create a conversation and append entries to it. Parameters:

Parameter	Type	Required	Default	Description
`agent_id`	string	No	—	Agent Id
`agent_version`	object	No	—	Agent Version
`completion_args`	object	No	—	White-listed arguments from the completion API
`description`	string	No	—	The description value
`guardrails`	object[]	No	—	The guardrails value
`handoff_execution`	string	No	—	Handoff Execution
`inputs`	object	Yes	—	The inputs value
`instructions`	string	No	—	The instructions value
`metadata`	object	No	—	Custom type for metadata with embedded validation.
`model`	string	No	—	The model value
`name`	string	No	—	The name value
`store`	boolean	No	—	The store value
`stream`	boolean	No	—	The stream value
`tools`	any[]	No	—	The tools value

Show inputSchema

{
  "type": "object",
  "properties": {
    "PCID": {
      "type": "string",
      "description": "Pink Connect ID for the authenticated connection"
    },
    "agent_id": {
      "type": "string",
      "description": "Agent Id"
    },
    "agent_version": {
      "description": "Agent Version"
    },
    "completion_args": {
      "type": "object",
      "description": "White-listed arguments from the completion API",
      "properties": {
        "stop": {
          "description": "The stop value"
        },
        "presence_penalty": {
          "type": "number",
          "description": "Presence Penalty"
        },
        "frequency_penalty": {
          "type": "number",
          "description": "Frequency Penalty"
        },
        "temperature": {
          "type": "number",
          "description": "The temperature value"
        },
        "top_p": {
          "type": "number",
          "description": "Top P"
        },
        "max_tokens": {
          "type": "integer",
          "description": "Max Tokens"
        },
        "random_seed": {
          "type": "integer",
          "description": "Random Seed"
        },
        "prediction": {
          "type": "object",
          "description": "Enable users to specify an expected completion, optimizing response times by leveraging known or predictable content."
        },
        "response_format": {
          "type": "object",
          "description": "Specify the format that the model must output. By default it will use `{ \"type\": \"text\" }`. Setting to `{ \"type\": \"json_object\" }` enables JSON mode, which guarantees the message the model generates is in JSON. When using JSON mode you MUST also instruct the model to produce JSON yourself with a system or a user message. Setting to `{ \"type\": \"json_schema\" }` enables JSON schema mode, which guarantees the message the model generates is in JSON and follows the schema you provide."
        },
        "tool_choice": {
          "type": "string",
          "description": "Tool Choice",
          "enum": [
            "auto",
            "none",
            "any",
            "required"
          ]
        },
        "reasoning_effort": {
          "type": "string",
          "description": "Controls the reasoning effort level for reasoning models. \"high\" enables comprehensive reasoning traces, \"none\" disables reasoning effort.",
          "enum": [
            "high",
            "none"
          ]
        }
      }
    },
    "description": {
      "type": "string",
      "description": "The description value"
    },
    "guardrails": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "block_on_error": {
            "type": "boolean",
            "description": "If true, return HTTP 403 and block request in the event of a server-side error"
          },
          "moderation_llm_v1": {
            "type": "object",
            "description": "Moderation Llm V1"
          },
          "moderation_llm_v2": {
            "type": "object",
            "description": "Moderation Llm V2"
          }
        }
      },
      "description": "The guardrails value"
    },
    "handoff_execution": {
      "type": "string",
      "description": "Handoff Execution",
      "enum": [
        "client",
        "server"
      ]
    },
    "inputs": {
      "description": "The inputs value"
    },
    "instructions": {
      "type": "string",
      "description": "The instructions value"
    },
    "metadata": {
      "type": "object",
      "description": "Custom type for metadata with embedded validation."
    },
    "model": {
      "type": "string",
      "description": "The model value"
    },
    "name": {
      "type": "string",
      "description": "The name value"
    },
    "store": {
      "type": "boolean",
      "description": "The store value"
    },
    "stream": {
      "type": "boolean",
      "description": "The stream value",
      "enum": [
        false
      ]
    },
    "tools": {
      "type": "array",
      "description": "The tools value"
    }
  },
  "required": [
    "PCID",
    "inputs"
  ]
}

​Tools

​mistral_agents_api_v1_agents_create

​mistral_agents_api_v1_agents_create_or_update_alias

​mistral_agents_api_v1_agents_delete

​mistral_agents_api_v1_agents_delete_alias

​mistral_agents_api_v1_agents_get

​mistral_agents_api_v1_agents_get_version

​mistral_agents_api_v1_agents_list

​mistral_agents_api_v1_agents_list_version_aliases

​mistral_agents_api_v1_agents_list_versions

​mistral_agents_api_v1_agents_update

​mistral_agents_api_v1_agents_update_version

​mistral_agents_api_v1_conversations_append

​mistral_agents_api_v1_conversations_delete

​mistral_agents_api_v1_conversations_get

​mistral_agents_api_v1_conversations_history

​mistral_agents_api_v1_conversations_list

​mistral_agents_api_v1_conversations_messages

​mistral_agents_api_v1_conversations_restart

​mistral_agents_api_v1_conversations_start

Tools

mistral_agents_api_v1_agents_create

mistral_agents_api_v1_agents_create_or_update_alias

mistral_agents_api_v1_agents_delete

mistral_agents_api_v1_agents_delete_alias

mistral_agents_api_v1_agents_get

mistral_agents_api_v1_agents_get_version

mistral_agents_api_v1_agents_list

mistral_agents_api_v1_agents_list_version_aliases

mistral_agents_api_v1_agents_list_versions

mistral_agents_api_v1_agents_update

mistral_agents_api_v1_agents_update_version

mistral_agents_api_v1_conversations_append

mistral_agents_api_v1_conversations_delete

mistral_agents_api_v1_conversations_get

mistral_agents_api_v1_conversations_history

mistral_agents_api_v1_conversations_list

mistral_agents_api_v1_conversations_messages

mistral_agents_api_v1_conversations_restart

mistral_agents_api_v1_conversations_start