mistral: AI Agent – Appmixer connector library

Basic Information

Name

appmixer.ai.mistral.AIAgent

Label

AI Agent

Version

1.0.0

Visibility

Public

Description

Build an AI agent on a Mistral model that answers with context and acts through tools. Connect a component to the 'tools' output port to let the agent call it as a tool, and mark the fields the model should fill with the 'Model Defined Parameter' variable; components connected behind it run as further steps of the same tool. For a tool with its own description and logic (branches, loops), connect a 'Tool Start' to the 'tools' port and end every branch with 'Tool Output'. An MCP server on the 'tools' port gives the agent all of its tools.

Properties

Field	Type	Description
Stream Responses	toggle	Stream partial responses as they arrive from the model. This only has an effect on the built-in chat utility; the agent still sends the complete answer to its output. For streaming to work, the 'Thread ID' of the agent must be set to the 'Thread ID' output of the 'Chat Trigger' component.

Input Ports

in

Field	Type	Description
Model	text	Insert a model ID or select one that supports function calling, for example `mistral-medium-latest`. Defaults to `mistral-medium-latest` when left empty.
Instructions	textarea	The system instructions the agent follows, for example 'You are a support agent. Look the order up with the tools before you answer.'. The maximum length is 256,000 characters.
Prompt	textarea	The message the agent responds to.
Thread ID	text	Set a thread ID to keep the context of the conversation: every message with the same ID continues the same thread. Without a thread ID the agent has no memory, which suits a simple question-and-answer step.
File	filepicker	An optional file the agent uses as additional context. PNG, JPEG, GIF and WEBP images go to the model directly (use a vision model), a PDF is read with Mistral OCR first, and other files are sent as text, which suits CSV, JSON or source code.
Memory Data Store	text	A data store that keeps the conversation history, so the context survives across executions and other flows can read or export it. Without it the history is kept in the component state.

Output Ports

out

Field	Type	Description
Answer	string	-
Prompt	string	-
Time (ms)	integer	-
Usage	object	-

tools

Field	Type	Description
Prompt	string	-
Model Defined Parameter	string	-

Full Metadata

#metadata-viewer { display: none; }{ "name": "appmixer.ai.mistral.AIAgent", "label": "AI Agent", "description": "Build an AI agent on a Mistral model that answers with context and acts through tools. Connect a component to the 'tools' output port to let the agent call it as a tool, and mark the fields the model should fill with the 'Model Defined Parameter' variable; components connected behind it run as further steps of the same tool. For a tool with its own description and logic (branches, loops), connect a 'Tool Start' to the 'tools' port and end every branch with 'Tool Output'. An MCP server on the 'tools' port gives the agent all of its tools.", "author": "Appmixer <info@appmixer.com>", "version": "1.0.0", "private": false, "auth": { "service": "appmixer:ai:mistral" }, "quota": { "manager": "appmixer:ai:mistral", "resources": "requests" }, "properties": { "schema": { "type": "object", "properties": { "stream": { "type": "boolean" } } }, "inspector": { "inputs": { "stream": { "type": "toggle", "label": "Stream Responses", "defaultValue": false, "index": 1, "tooltip": "Stream partial responses as they arrive from the model. This only has an effect on the built-in chat utility; the agent still sends the complete answer to its output. For streaming to work, the 'Thread ID' of the agent must be set to the 'Thread ID' output of the 'Chat Trigger' component." } } } }, "inPorts": [ { "name": "in", "schema": { "type": "object", "properties": { "model": { "type": "string" }, "instructions": { "type": "string", "maxLength": 256000 }, "prompt": { "type": "string" }, "threadId": { "type": "string" }, "fileId": { "type": "string" }, "storeId": { "type": "string" } }, "required": [ "prompt" ] }, "inspector": { "inputs": { "model": { "type": "text", "index": 1, "label": "Model", "defaultValue": "mistral-medium-latest", "tooltip": "Insert a model ID or select one that supports function calling, for example <code>mistral-medium-latest</code>. Defaults to <code>mistral-medium-latest</code> when left empty.", "source": { "url": "/component/appmixer/ai/mistral/ListModels?outPort=out", "data": { "properties": { "isSource": true }, "transform": "./ListModels#toAgentModelSelectArray" } } }, "instructions": { "type": "textarea", "index": 2, "label": "Instructions", "tooltip": "The system instructions the agent follows, for example 'You are a support agent. Look the order up with the tools before you answer.'. The maximum length is 256,000 characters." }, "prompt": { "type": "textarea", "index": 3, "label": "Prompt", "tooltip": "The message the agent responds to." }, "threadId": { "type": "text", "index": 4, "label": "Thread ID", "tooltip": "Set a thread ID to keep the context of the conversation: every message with the same ID continues the same thread. Without a thread ID the agent has no memory, which suits a simple question-and-answer step." }, "fileId": { "type": "filepicker", "index": 5, "label": "File", "tooltip": "An optional file the agent uses as additional context. PNG, JPEG, GIF and WEBP images go to the model directly (use a vision model), a PDF is read with Mistral OCR first, and other files are sent as text, which suits CSV, JSON or source code." }, "storeId": { "type": "text", "index": 6, "label": "Memory Data Store", "source": { "url": "/component/appmixer/utils/storage/ListStores?outPort=out", "data": { "transform": "./ListStores#toSelectArray" } }, "tooltip": "A data store that keeps the conversation history, so the context survives across executions and other flows can read or export it. Without it the history is kept in the component state." } } } } ], "outPorts": [ { "name": "out", "schema": { "type": "object", "properties": { "answer": { "type": "string", "title": "Answer", "example": "Your order 1042 shipped yesterday and arrives on Friday." }, "prompt": { "type": "string", "title": "Prompt", "example": "Where is my order 1042?" }, "time": { "type": "integer", "title": "Time (ms)", "example": 1834 }, "usage": { "type": "object", "title": "Usage", "properties": { "prompt_tokens": { "type": "integer", "title": "Usage.Prompt Tokens", "example": 412 }, "completion_tokens": { "type": "integer", "title": "Usage.Completion Tokens", "example": 38 }, "total_tokens": { "type": "integer", "title": "Usage.Total Tokens", "example": 450 } }, "example": { "prompt_tokens": 412, "completion_tokens": 38, "total_tokens": 450 } } }, "required": [ "answer", "prompt", "time" ] } }, { "name": "tools", "schema": { "type": "object", "properties": { "prompt": { "type": "string", "title": "Prompt", "example": "What is the weather in Paris?" }, "modelDefinedParameter": { "type": "string", "title": "Model Defined Parameter", "example": "Paris" } } } } ], "icon": "data:image/svg+xml;base64,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" }