# Mistral AI

Mistral AI is a research lab building state-of-the-art open-source language models and providing APIs for developers and enterprises to integrate these models into their applications.

- **Category:** artificial intelligence
- **Auth:** API_KEY
- **Composio Managed App Available?** N/A
- **Tools:** 54
- **Triggers:** 0
- **Slug:** `MISTRAL_AI`
- **Version:** 20260307_00

## Tools

### Append to conversation

**Slug:** `MISTRAL_AI_APPEND_TO_CONVERSATION`

Tool to append new entries to an existing conversation in Mistral AI. Use when you need to continue a conversation by adding new messages or inputs. This is a beta endpoint.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `store` | boolean | No | Whether to persist the conversation results to the server. Defaults to true |
| `inputs` | string | Yes | New input to append to the conversation. Can be a string (simple text input) or an array of entry objects (MessageInputEntry, MessageOutputEntry, FunctionResultEntry, etc.) |
| `completion_args` | object | No | White-listed arguments from the completion API to configure the response generation (e.g., temperature, max_tokens) |
| `conversation_id` | string | Yes | ID of the conversation to which entries will be appended |
| `handoff_execution` | string ("client" | "server") | No | Handoff execution mode for agent handoffs. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create Agent

**Slug:** `MISTRAL_AI_CREATE_AGENT`

Tool to create a new AI agent with custom configuration (Beta). Use when you need to create an agent with specific model, name, instructions, and tools.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `name` | string | Yes | Name of the agent |
| `model` | string | Yes | Model ID to use for the agent (e.g., 'mistral-large-latest', 'mistral-small-latest') |
| `tools` | array | No | List of tools available to the agent. Can include function tools, web search, code interpreter, image generation, or document library tools |
| `handoffs` | array | No | Optional handoff configuration for agent-to-agent transfers |
| `metadata` | object | No | Optional metadata to attach to the agent |
| `description` | string | No | Optional description of the agent's purpose and capabilities |
| `instructions` | string | No | System instructions that guide the agent's behavior during conversations |
| `completion_args` | object | No | White-listed arguments from the completion API. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create Agents Completion

**Slug:** `MISTRAL_AI_CREATE_AGENTS_COMPLETION`

Tool to generate completions using a Mistral AI agent with specific instructions and tools. Use when you need an agent to process messages and generate responses. Agents can use tools, follow instructions, and maintain conversation context.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `n` | integer | No | Number of completions to generate. Input tokens are billed once |
| `stop` | string | No | Stop generation when this token (or any of these tokens if array) is detected |
| `tools` | array | No | List of tools available for the agent to use during completion |
| `top_p` | number | No | Nucleus sampling parameter (0.0-1.0). Consider only tokens with cumulative probability >= top_p |
| `stream` | boolean | No | Whether to stream partial responses. If true, returns server-sent events. Default false |
| `agent_id` | string | Yes | ID of the agent to use for this completion. Obtain from listing or creating agents |
| `messages` | array | Yes | List of messages forming the conversation. Must contain at least one message. Each message has a role ('system', 'user', 'assistant', 'tool') and content |
| `metadata` | object | No | Additional metadata to attach to the completion request |
| `max_tokens` | integer | No | Maximum number of tokens to generate. The total of prompt + max_tokens cannot exceed model's context length |
| `prediction` | object | No | Prediction configuration for optimizing response times. |
| `prompt_mode` | string | No | Prompt mode: 'reasoning' to enable reasoning mode with system prompt for reasoning models |
| `random_seed` | integer | No | Random seed for deterministic sampling. Same seed with same inputs produces same outputs |
| `temperature` | number | No | Sampling temperature (0.0-2.0). Higher values make output more random, lower values more deterministic |
| `tool_choice` | string | No | Controls tool selection: 'auto' (default), 'any', 'none', or specific tool object |
| `response_format` | object | No | Response format configuration. |
| `presence_penalty` | number | No | Penalty for token presence (-2.0 to 2.0). Positive values encourage diverse vocabulary |
| `frequency_penalty` | number | No | Penalty for token frequency (-2.0 to 2.0). Positive values discourage repetition based on frequency |
| `parallel_tool_calls` | boolean | No | Whether to allow parallel tool calls. Default true |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create Audio Transcription

**Slug:** `MISTRAL_AI_CREATE_AUDIO_TRANSCRIPTION`

Transcribe audio files to text using Mistral AI's Voxtral models. Use this action to convert speech in audio files to written text. Supports multiple input methods: file upload, file_id from previously uploaded files, or file_url for publicly accessible audio. The transcription supports 13 languages with automatic language detection, speaker diarization, and configurable timestamp granularities (segment or word level). Key features: - Multi-language support (English, Chinese, Hindi, Spanish, Arabic, French, Portuguese, Russian, German, Japanese, Korean, Italian, Dutch) - Speaker diarization to identify different speakers - Word and segment-level timestamps - Context biasing for domain-specific terminology - Supports common audio formats (mp3, wav, m4a, etc.) Note: Exactly one of file, file_id, or file_url must be provided. The action does not support streaming mode.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file` | object | No | Audio file to transcribe. Supported formats include common audio formats (mp3, wav, m4a, etc.). Mutually exclusive with file_id and file_url - provide only one of these three options. |
| `model` | string | Yes | ID of the model to use for transcription. Use 'voxtral-mini-latest' for the latest version or 'voxtral-mini-2602' for a specific version. Supports 13 languages: English, Chinese, Hindi, Spanish, Arabic, French, Portuguese, Russian, German, Japanese, Korean, Italian, and Dutch. |
| `diarize` | boolean | No | Enable speaker diarization to identify different speakers in the audio. When enabled, the response will include speaker labels with timestamps. |
| `file_id` | string | No | ID of a previously uploaded file to transcribe. Obtain this from MISTRAL_AI_UPLOAD_FILE or MISTRAL_AI_LIST_FILES actions. Mutually exclusive with file and file_url - provide only one of these three options. |
| `file_url` | string | No | URL of an audio file to transcribe. The file must be publicly accessible. Mutually exclusive with file and file_id - provide only one of these three options. |
| `language` | string | No | Language code of the audio (e.g., 'en' for English, 'fr' for French, 'es' for Spanish, 'zh' for Chinese). Providing the language can improve transcription accuracy. If not provided, the language will be auto-detected. |
| `temperature` | number | No | Sampling temperature between 0 and 1. Higher values (e.g., 0.8) make output more random, lower values (e.g., 0.2) make it more focused and deterministic. |
| `context_bias` | array | No | List of words or phrases to bias the transcription towards. Each item should be a single word or phrase without commas or whitespace within it. Use this to improve accuracy for domain-specific terminology. |
| `timestamp_granularities` | array | No | Granularities of timestamps to include in the response. Options: 'segment' for segment-level timestamps, 'word' for word-level timestamps. Can include both for detailed timing information. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create Chat Completion

**Slug:** `MISTRAL_AI_CREATE_CHAT_COMPLETION`

Generate conversational responses from Mistral AI models. Supports streaming, function calling, and various model parameters. Use when you need to create chat completions with Mistral AI models for conversation, question answering, or function calling scenarios.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `n` | integer | No | Number of completions to return for each request. Input tokens are only billed once. Useful for generating multiple variations. |
| `stop` | string | No | Stop generation if this token/string is detected. Can be a single string or list of strings. |
| `model` | string | Yes | ID of the model to use. Use the List Available Models API to see available models (e.g., 'mistral-tiny', 'mistral-small', 'mistral-medium', 'mistral-large-latest'). |
| `tools` | array | No | List of tools (functions) the model may call. Each tool should have 'type', 'function' with 'name', 'description', and 'parameters' (JSON schema). |
| `top_p` | number | No | Nucleus sampling probability mass. The model considers tokens with top_p probability mass. Default 1.0. We recommend altering this or temperature but not both. |
| `stream` | boolean | No | Whether to stream back partial progress as server-sent events. When true, tokens are sent as they become available. Default false. |
| `messages` | array | Yes | List of messages in the conversation. Each message has a 'role' (system/user/assistant/tool) and 'content'. |
| `metadata` | object | No | Optional metadata to include with the request for tracking or logging purposes. |
| `max_tokens` | integer | No | Maximum number of tokens to generate in the completion. Prompt tokens plus max_tokens cannot exceed the model's context length. |
| `prediction` | object | No | Expected completion for optimizing response times. |
| `prompt_mode` | string | No | Toggle between reasoning mode and no system prompt. When set to 'reasoning', the system prompt for reasoning models will be used. |
| `random_seed` | integer | No | Seed for random sampling. If set, different calls will generate deterministic results (same input = same output). |
| `safe_prompt` | boolean | No | Whether to inject a safety prompt before all conversations to reduce harmful outputs. Default false. |
| `temperature` | number | No | Sampling temperature between 0.0 and 1.0. Higher values (e.g., 0.7) make output more random, lower values (e.g., 0.2) make it more focused and deterministic. Recommended range: 0.0-0.7. |
| `tool_choice` | string | No | Controls which tool is called by the model. 'none' means no tool calls, 'auto' lets model decide, 'any'/'required' means must call a tool. Or specify a tool: {"type": "function", "function": {"name": "my_function"}}. Default 'auto'. |
| `response_format` | object | No | Format specification for model output. |
| `presence_penalty` | number | No | Penalizes word/phrase repetition regardless of frequency. Range: -2.0 to 2.0. Higher values encourage more diverse vocabulary. Default 0.0. |
| `frequency_penalty` | number | No | Penalizes word repetition based on frequency in the generated text. Range: -2.0 to 2.0. Higher values discourage repetition. Default 0.0. |
| `parallel_tool_calls` | boolean | No | Whether to enable parallel function calling during tool use. When enabled, the model can call multiple tools simultaneously. Default true. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create Chat Moderation

**Slug:** `MISTRAL_AI_CREATE_CHAT_MODERATION`

Tool to classify chat content for moderation purposes across 9 categories. Use when you need to detect harmful content, inappropriate messages, or policy violations in chat conversations.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `input` | array | Yes | Array of message objects to classify. Each message has a 'role' (user/assistant) and 'content' (string). The last user message will be classified for moderation. |
| `model` | string | No | Moderation model to use for classification. Use 'mistral-moderation-latest' for the most recent version. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create Embeddings

**Slug:** `MISTRAL_AI_CREATE_EMBEDDINGS`

Tool to generate vector embeddings for input text using Mistral AI embedding models. Use when you need to convert text into numerical vectors for semantic search, similarity comparison, or RAG applications.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `input` | string | Yes | The text content to be embedded. Can be a single string or an array of strings for batch processing. |
| `model` | string | Yes | The ID of the model to be used for embedding (e.g., 'mistral-embed'). |
| `output_dtype` | string ("float" | "int8" | "uint8" | "binary" | "ubinary") | No | Output data type for embeddings. |
| `encoding_format` | string ("float" | "base64") | No | Output encoding format for embeddings. |
| `output_dimension` | integer | No | The dimension of the output embeddings when feature available. If not provided, a default output dimension will be used. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create FIM Completion

**Slug:** `MISTRAL_AI_CREATE_FIM_COMPLETION`

Generate code completions using fill-in-the-middle functionality. Use when you need to complete code between a prefix and suffix, or continue code from a prompt. Ideal for code completion, function implementation, and context-aware code generation.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `stop` | string | No | Stop generation if this token is detected. Can be a single string or an array of strings. When provided as array, generation stops if any of the tokens is detected. |
| `model` | string | No | ID of the model with FIM capability to use. Default: codestral-latest. |
| `top_p` | number | No | Nucleus sampling probability mass. The model considers tokens with top_p probability mass. For example, 0.1 means only tokens comprising the top 10% probability mass are considered. Default 1.0. Generally recommend altering this or temperature but not both. |
| `prompt` | string | Yes | The text/code to complete. This is the prefix that comes before the gap to be filled. |
| `stream` | boolean | No | This parameter is not supported and will always be set to false. Streaming responses are not handled by this action. |
| `suffix` | string | No | Optional text/code that adds more context for the model. When given both prompt and suffix, the model fills what is between them. When suffix is not provided, the model simply executes completion starting with prompt. |
| `max_tokens` | integer | No | Maximum number of tokens to generate in the completion. The token count of prompt plus max_tokens cannot exceed the model's context length. |
| `random_seed` | integer | No | Seed for random sampling. If set, different calls will generate deterministic results. Use for reproducible completions. |
| `temperature` | number | No | Sampling temperature between 0.0 and 1.0. Higher values like 0.7 make output more random, lower values like 0.2 make it more focused and deterministic. Recommended range: 0.0-0.7. Generally recommend altering this or top_p but not both. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create library

**Slug:** `MISTRAL_AI_CREATE_LIBRARY`

Tool to create a new document library. Use when you need to group documents into a new library. Use after confirming authentication.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `name` | string | Yes | Name of the new document library |
| `description` | string | No | Optional description of the new document library |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create library share

**Slug:** `MISTRAL_AI_CREATE_LIBRARY_SHARE`

Create or update sharing permissions for a library. Use to grant access to users, workspaces, or organizations. Specify the access level (Viewer or Editor) and the entity to share with. This is a beta endpoint.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `level` | string ("Viewer" | "Editor") | Yes | Access level to grant: 'Viewer' for read-only access or 'Editor' for read and write access |
| `org_id` | string | No | Optional organization ID to associate with the share (UUID format) |
| `library_id` | string | Yes | Unique identifier of the library to share (UUID format) |
| `share_with_type` | string ("User" | "Workspace" | "Org") | Yes | The type of entity to share the library with: 'User' for individual users, 'Workspace' for team workspaces, or 'Org' for entire organizations |
| `share_with_uuid` | string | Yes | The UUID of the entity (user, workspace, or organization) to share the library with |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create Moderation

**Slug:** `MISTRAL_AI_CREATE_MODERATION`

Tool to classify text content for moderation purposes across 9 categories. Use when you need to detect harmful content, inappropriate text, or policy violations in raw text inputs.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `input` | string | Yes | Text to classify for moderation. Can be a single string or an array of strings for batch processing. |
| `model` | string | No | Moderation model to use for classification. Use 'mistral-moderation-latest' for the most recent version. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create OCR

**Slug:** `MISTRAL_AI_CREATE_OCR`

Extract text and structured data from images and documents using Mistral AI's OCR capabilities. Supports PDFs, images, tables, headers, footers, and custom structured extraction. Use when you need to digitize documents, extract invoice data, parse forms, or convert scanned documents to text.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `id` | string | No | Optional custom identifier for this OCR request. |
| `model` | string | Yes | Model to use for OCR processing. Use 'mistral-ocr-latest' for the latest version or specific version like 'mistral-ocr-2512'. |
| `pages` | array | No | Specific pages to process (0-indexed). Can be single numbers, ranges, or lists. If not specified, all pages are processed. |
| `document` | string | Yes | Document to run OCR on. Can be an image URL, document URL, or file ID of a previously uploaded file. |
| `image_limit` | integer | No | Maximum number of images to extract from the document. |
| `table_format` | string ("markdown" | "html") | No | Format for extracted tables. |
| `extract_footer` | boolean | No | Whether to extract footer content from the document. Default false. |
| `extract_header` | boolean | No | Whether to extract header content from the document. Default false. |
| `image_min_size` | integer | No | Minimum height and width (in pixels) for images to extract. Smaller images are ignored. |
| `include_image_base64` | boolean | No | Whether to include base64-encoded image data in the response. |
| `bbox_annotation_format` | object | No | Response format specification for structured output. |
| `document_annotation_format` | object | No | Response format specification for structured output. |
| `document_annotation_prompt` | string | No | Optional prompt to guide the model in extracting structured output from the entire document. Requires document_annotation_format to be provided. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Create or Update Agent Alias

**Slug:** `MISTRAL_AI_CREATE_OR_UPDATE_AGENT_ALIAS`

Tool to create or update an agent version alias. Use when you need to assign a version alias (like 'production' or 'staging') to a specific agent version.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `alias` | string | Yes | Name of the version alias (e.g., 'production', 'staging', 'latest'). |
| `version` | integer | Yes | Agent version number to assign to this alias. Must be a non-negative integer. |
| `agent_id` | string | Yes | Unique identifier of the agent (UUID format). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Delete agent

**Slug:** `MISTRAL_AI_DELETE_AGENT`

Permanently deletes an agent by its ID (Beta feature). Use this tool when you need to remove an agent that is no longer needed. This operation is irreversible - the agent will be permanently removed from the system.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `agent_id` | string | Yes | The unique identifier of the agent to delete. This should be the agent ID returned from previous agent operations (e.g., ag_019c5ecefcc37282b0fff41b4b3a8f5d). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Delete Conversation

**Slug:** `MISTRAL_AI_DELETE_CONVERSATION`

Tool to delete a conversation by its ID (Beta). Use when you need to permanently remove a conversation. This is a beta feature.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `conversation_id` | string | Yes | ID of the conversation to delete. Must be a valid conversation ID. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Delete File

**Slug:** `MISTRAL_AI_DELETE_FILE`

Delete a file by its ID from Mistral AI. Permanently removes the file and its metadata. Use List Files action first to obtain valid file IDs.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file_id` | string | Yes | UUID of the file to delete. Obtain file IDs from the List Files action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Delete library

**Slug:** `MISTRAL_AI_DELETE_LIBRARY`

Permanently deletes a library and all of its documents from Mistral AI. Use this tool when you need to remove an entire library including all its documents. This operation is irreversible - the library and all its documents will be permanently removed. Returns the deleted library details on success. Common errors: - 404: Library not found (already deleted or invalid library_id) - 422: Invalid UUID format for library_id

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | The unique UUID identifier of the library to delete. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Delete library document

**Slug:** `MISTRAL_AI_DELETE_LIBRARY_DOCUMENT`

Permanently deletes a document from a Mistral AI library. Use this tool when you need to remove a specific document from a library. Both library_id and document_id must be valid UUIDs. This operation is irreversible - the document will be permanently removed. Returns an empty response on success (HTTP 204). Common errors: - 404: Document not found (already deleted or invalid document_id) - 422: Invalid UUID format for library_id or document_id

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | The unique UUID identifier of the library containing the document. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |
| `document_id` | string | Yes | The unique UUID identifier of the document to delete. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Delete library share

**Slug:** `MISTRAL_AI_DELETE_LIBRARY_SHARE`

Remove sharing permissions for a library from a user, workspace, or organization. Use when you need to revoke access to a library that was previously shared. This is a beta feature.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `org_id` | string | No | Optional organization ID for scoping the operation. |
| `library_id` | string | Yes | The unique UUID identifier of the library to remove sharing permissions from. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |
| `share_with_type` | string ("User" | "Workspace" | "Org") | Yes | The type of entity (User, Workspace, or Org) to revoke access from. |
| `share_with_uuid` | string | Yes | The UUID of the entity (user, workspace or organization) to revoke access from. Must be a valid UUID format. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Download File

**Slug:** `MISTRAL_AI_DOWNLOAD_FILE`

Download the content of a previously uploaded file from Mistral AI. Returns the raw binary content of the file. Use this when you need to retrieve file data for processing, such as training data files (.jsonl) for fine-tuning. Requires a valid file_id which can be obtained from the list_files or upload_file actions.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file_id` | string | Yes | The unique identifier (UUID) of the file to download. Obtain this from the list_files or upload_file actions. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Agent

**Slug:** `MISTRAL_AI_GET_AGENT`

Tool to retrieve details of a specific Mistral AI agent by its ID. Returns comprehensive agent information including model, instructions, tools, and configuration. Use when you need to inspect or verify an agent's settings.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `agent_id` | string | Yes | UUID of the agent to retrieve. Obtain agent IDs from the List Agents action. |
| `agent_version` | string | No | Specific version of the agent to retrieve. If not provided, the latest version is returned. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Agent Version

**Slug:** `MISTRAL_AI_GET_AGENT_VERSION`

Retrieve a specific version of an agent (Beta). Use when you need to get details about a particular agent version, including its configuration, tools, and metadata.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `version` | string | Yes | Version number of the agent to retrieve (e.g., '0', '1', '2'). |
| `agent_id` | string | Yes | UUID of the agent to retrieve. Obtain agent IDs from the List Agents action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Conversation

**Slug:** `MISTRAL_AI_GET_CONVERSATION`

Tool to retrieve details of a specific conversation. Use when you need to fetch conversation metadata including timestamps, configuration, and associated model or agent information.

#### Input Parameters

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

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Conversation History

**Slug:** `MISTRAL_AI_GET_CONVERSATION_HISTORY`

Retrieve the full history of a conversation in Mistral AI. Returns all entries including messages, tool calls, function results, and agent handoffs. Use this to review conversation context or export conversation data. Note: This is a beta endpoint.

#### Input Parameters

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

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Conversation Messages

**Slug:** `MISTRAL_AI_GET_CONVERSATION_MESSAGES`

Retrieve all messages from a Mistral AI conversation. Use when you need to fetch the complete message history for a specific conversation.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `conversation_id` | string | Yes | ID of the conversation from which to fetch messages. This is a unique identifier for the conversation in Mistral AI's system. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get document extracted text URL

**Slug:** `MISTRAL_AI_GET_DOCUMENT_EXTRACTED_TEXT_URL`

Retrieve a signed URL to download the extracted text from a document in a Mistral AI library. This is a beta endpoint. Note: Only documents that undergo OCR processing (such as PDFs) will have extracted text available; plain text files that don't require OCR will return a 404 error.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | The unique UUID identifier of the library containing the document. Use MISTRAL_AI_LIST_LIBRARIES to get available library IDs. |
| `document_id` | string | Yes | The unique UUID identifier of the document. Use MISTRAL_AI_LIST_LIBRARY_DOCUMENTS to get document IDs. Note: Only documents that undergo OCR processing (e.g., PDFs) will have extracted text available. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get document signed URL

**Slug:** `MISTRAL_AI_GET_DOCUMENT_SIGNED_URL`

Get a signed URL to download a document from a Mistral AI library. Returns a temporary URL that provides direct access to download the document content. Use this when you need to retrieve document files from a library. The signed URL is typically valid for 30 minutes.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | The unique UUID identifier of the library containing the document. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |
| `document_id` | string | Yes | The unique UUID identifier of the document to retrieve the signed URL for. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Document Status

**Slug:** `MISTRAL_AI_GET_DOCUMENT_STATUS`

Retrieve the processing status of a document in a Mistral AI library. Use this to check if a document has finished processing after upload. Returns the document ID and its current processing status.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | The unique identifier (UUID) for the document library. Obtain this from the List Libraries action or from the library URL in Le Chat. |
| `document_id` | string | Yes | The unique identifier (UUID) for the document. Obtain this from the List Library Documents action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Document Text Content

**Slug:** `MISTRAL_AI_GET_DOCUMENT_TEXT_CONTENT`

Retrieve the extracted text content of a specific document from a Mistral AI library (Beta). Returns the full text content extracted from the document. Use the List Libraries action first to obtain valid library IDs, then use List Library Documents to get document IDs.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | The unique identifier (UUID) for the document library. Obtain this from the List Libraries action or from the library URL in Le Chat. |
| `document_id` | string | Yes | The unique identifier (UUID) for the document. Obtain this from the List Library Documents action or from the document URL in Le Chat. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get File Signed URL

**Slug:** `MISTRAL_AI_GET_FILE_SIGNED_URL`

Get a time-limited signed URL for downloading a file from Mistral AI. Use when you need a temporary download link that can be shared or used externally. The URL expires after the specified number of hours (default 24).

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `expiry` | integer | No | Number of hours before the URL becomes invalid. Defaults to 24 hours if not specified. |
| `file_id` | string | Yes | UUID of the file to get signed URL for. Obtain file IDs from the List Files action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Fine Tuning Jobs

**Slug:** `MISTRAL_AI_GET_FINE_TUNING_JOBS`

List fine-tuning jobs with optional filtering and pagination. Use this tool to retrieve all fine-tuning jobs for your organization. Supports filtering by model, status, creation time, and W&B integration. Results are paginated; use 'page' and 'page_size' to navigate large result sets.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `page` | integer | No | Page number of results to return (0-indexed). Default is 0. |
| `model` | string | No | Filter by the base model name used for fine-tuning. When set, only jobs using this model are returned. |
| `status` | string | No | Filter by the current job status. Valid values: QUEUED, STARTED, VALIDATING, VALIDATED, RUNNING, FAILED_VALIDATION, FAILED, SUCCESS, CANCELLED, CANCELLATION_REQUESTED. |
| `suffix` | string | No | Filter by the model suffix used when creating the fine-tuned model. |
| `page_size` | integer | No | Number of fine-tuning jobs to return per page (1-1000). Default is 100. |
| `wandb_name` | string | No | Filter by the Weights and Biases run name. |
| `created_after` | string | No | Filter to return only jobs created after this date/time (ISO 8601 format). |
| `created_by_me` | boolean | No | When set to true, only return jobs created by the API caller. Default is false. |
| `wandb_project` | string | No | Filter by the Weights and Biases project name. |
| `created_before` | string | No | Filter to return only jobs created before this date/time (ISO 8601 format). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get library

**Slug:** `MISTRAL_AI_GET_LIBRARY`

Retrieve detailed information about a specific library. Returns complete library metadata including name, description, document counts, size, timestamps, and ownership details. Use List Libraries action first to obtain valid library IDs.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | UUID of the library to retrieve. Obtain library IDs from the List Libraries action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Library Document

**Slug:** `MISTRAL_AI_GET_LIBRARY_DOCUMENT`

Retrieve metadata for a specific document in a Mistral AI library. Returns detailed information including processing status, size, summary, token counts, and timestamps. Use this to check document status after upload or to retrieve details for a known document.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | The unique identifier (UUID) for the document library. Obtain this from the List Libraries action or from the library URL in Le Chat. |
| `document_id` | string | Yes | The unique identifier (UUID) for the document. Obtain this from the List Library Documents action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Get Model

**Slug:** `MISTRAL_AI_GET_MODEL`

Tool to retrieve detailed information about a specific Mistral AI model by its ID. Returns model metadata including capabilities, context length, and ownership. Use when you need to inspect model specifications before using it.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `model_id` | string | Yes | The ID of the model to retrieve. Use model identifiers like 'mistral-small-latest', 'mistral-large-latest', or fine-tuned model IDs in format 'ft:open-mistral-7b:587a6b29:20240514:7e773925'. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List agent aliases

**Slug:** `MISTRAL_AI_LIST_AGENT_ALIASES`

Retrieve all aliases for an agent version. Use to view and manage version aliases for an agent. Note: This is a beta endpoint.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `agent_id` | string | Yes | Unique identifier of the agent. Obtain agent IDs from the List Agents action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Agents

**Slug:** `MISTRAL_AI_LIST_AGENTS`

Tool to list all configured agents (Beta). Use when you need to retrieve a list of agents available in your organization.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `id` | string | No | Filter agents by ID. |
| `name` | string | No | Filter agents by name. |
| `page` | integer | No | Page number (0-indexed) for pagination. Default is 0. |
| `sources` | string | No | Filter agents by source. |
| `metadata` | string | No | Filter agents by metadata. |
| `page_size` | integer | No | Number of agents to return per page. Default is 20. |
| `deployment_chat` | boolean | No | Filter agents by deployment chat setting. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Agent Versions

**Slug:** `MISTRAL_AI_LIST_AGENT_VERSIONS`

List all versions of a specific agent. Use when you need to view the version history of an agent. Note: This is a beta endpoint.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `page` | integer | No | Page number for pagination (0-indexed). Default is 0. |
| `agent_id` | string | Yes | Unique identifier of the agent to retrieve versions for. |
| `page_size` | integer | No | Number of versions to return per page. Default is 20. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Batch Jobs

**Slug:** `MISTRAL_AI_LIST_BATCH_JOBS`

Tool to retrieve a list of all batch jobs with optional filtering and pagination. Use when you need to view or manage batch processing jobs.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `page` | integer | No | Page number of results to return (0-indexed). Default is 0. |
| `model` | string | No | Filter by the model name used for batch processing. When set, only jobs using this model are returned. |
| `status` | string | No | Filter by the current batch job status. Valid values include: QUEUED, RUNNING, SUCCESS, FAILED, CANCELLED. |
| `agent_id` | string | No | Filter by agent ID. When set, only batch jobs associated with this agent are returned. |
| `metadata` | string | No | Filter by metadata string. When set, only batch jobs with matching metadata are returned. |
| `page_size` | integer | No | Number of batch jobs to return per page (1-1000). Default is 100. |
| `created_after` | string | No | Filter to return only jobs created after this date/time (ISO 8601 format). |
| `created_by_me` | boolean | No | When set to true, only return batch jobs created by the API caller. Default is false. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Conversations

**Slug:** `MISTRAL_AI_LIST_CONVERSATIONS`

List all created conversations (Beta). Use to retrieve conversation history or manage existing conversations. Supports pagination and metadata filtering.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `page` | integer | No | Page number of results to return (0-indexed). Default is 0. |
| `metadata` | string | No | Filter conversations by metadata. Exact format depends on your metadata structure. |
| `page_size` | integer | No | Number of conversations to return per page (1-1000). Default is 100. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Files

**Slug:** `MISTRAL_AI_LIST_FILES`

Tool to list all files available to the user. Use when you need to view or manage uploaded files, supports pagination.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `after` | string | No | Pagination cursor: file ID to start after. Use to retrieve the next page of results. |
| `limit` | integer | No | Maximum number of files to return, must be >=1. Default is server-side limit. |
| `order` | string ("asc" | "desc") | No | Order to list files: 'asc' for ascending or 'desc' for descending. Default 'desc'. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List libraries

**Slug:** `MISTRAL_AI_LIST_LIBRARIES`

List all document libraries accessible to your organization. Returns library metadata including id, name, description, document counts, and timestamps. Use to discover available libraries before listing or uploading documents. Note: This is a beta endpoint.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `limit` | integer | No | (Beta) Maximum number of libraries to return per page. May not be honored in beta. |
| `page_token` | string | No | (Beta) Token for fetching the next page of results. May not be honored in beta. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Library Documents

**Slug:** `MISTRAL_AI_LIST_LIBRARY_DOCUMENTS`

List all documents in a Mistral AI document library. Returns document metadata including name, processing status, size, summary, and timestamps. Use the List Libraries action first to obtain valid library IDs. Supports pagination for large libraries.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `page` | integer | No | Page number to retrieve (0-indexed). Use with page_size for pagination. |
| `page_size` | integer | No | Number of documents per page. Defaults to 100 if not specified. |
| `library_id` | string | Yes | The unique identifier (UUID) for the document library. Obtain this from the List Libraries action or from the library URL in Le Chat. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List library shares

**Slug:** `MISTRAL_AI_LIST_LIBRARY_SHARES`

List all sharing permissions for a document library. Returns details about who has access to the library, including role, share type, and user/organization identifiers. Use the List Libraries action first to obtain valid library IDs. Note: This is a beta endpoint.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | The unique identifier (UUID) for the document library. Obtain this from the List Libraries action or from the library URL in Le Chat. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### List Models

**Slug:** `MISTRAL_AI_LIST_MODELS`

Tool to retrieve all available Mistral AI models including base models and fine-tuned models. Use when you need to see what models are available for chat completions, embeddings, or fine-tuning.

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Reprocess document

**Slug:** `MISTRAL_AI_REPROCESS_DOCUMENT`

Reprocess a document in a Mistral AI library (Beta). Use when you need to trigger reprocessing of a document, such as after updating library settings or to refresh document embeddings. Both library_id and document_id must be valid UUIDs. Returns an empty response on success (HTTP 204).

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `library_id` | string | Yes | The unique UUID identifier of the library containing the document. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |
| `document_id` | string | Yes | The unique UUID identifier of the document to reprocess. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Restart Conversation

**Slug:** `MISTRAL_AI_RESTART_CONVERSATION`

Tool to restart a conversation from a specific point (Beta). Use when you need to branch a conversation or replay it from a particular message. Creates a new conversation starting from the specified entry.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `store` | boolean | No | Whether to persist the conversation results to the server. Defaults to true |
| `inputs` | string | Yes | New input to append when restarting the conversation. Can be a string (simple text input) or an array of entry objects (MessageInputEntry, MessageOutputEntry, FunctionResultEntry, etc.) |
| `stream` | boolean | No | Enable response streaming for real-time output. Defaults to false |
| `metadata` | object | No | Additional metadata to associate with the restarted conversation |
| `agent_version` | string | No | Version of the agent to use when restarting the conversation |
| `from_entry_id` | string | Yes | ID of the entry from which to restart the conversation. The conversation will be recreated from this point |
| `completion_args` | object | No | White-listed arguments from the completion API to configure the response generation (e.g., temperature, max_tokens) |
| `conversation_id` | string | Yes | ID of the original conversation which is being restarted |
| `handoff_execution` | string ("client" | "server") | No | Handoff execution mode for agent handoffs. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Retrieve File

**Slug:** `MISTRAL_AI_RETRIEVE_FILE`

Retrieve metadata of a file uploaded to Mistral AI. Returns file details including size, filename, purpose, and creation time. Use List Files action first to obtain valid file IDs.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file_id` | string | Yes | UUID of the file to retrieve. Obtain file IDs from the List Files action. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Start Conversation

**Slug:** `MISTRAL_AI_START_CONVERSATION`

Tool to start a new conversation with a Mistral AI agent or base model. Use when initiating a conversational interaction that requires context tracking. Either 'model' or 'agent_id' must be provided. Returns a conversation_id for continuing the conversation.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `name` | string | No | Name for the conversation to help identify it later. |
| `model` | string | No | Specifies which base model to use. Either 'model' or 'agent_id' must be set. Examples: 'mistral-small-latest', 'mistral-large-latest', 'mistral-medium-latest'. |
| `store` | boolean | No | Whether to persist the conversation history. Set to false to prevent cloud storage. Default is true. |
| `tools` | array | No | Available tools for the conversation (FunctionTool, WebSearchTool, CodeInterpreterTool, etc.). Array of tool definition objects. |
| `inputs` | string | Yes | The initial message(s) to append to the conversation. Can be a simple string or an array of entry objects with 'content' and 'role' fields. |
| `stream` | boolean | No | Set true for streaming responses as Server-Sent Events. Default is false (returns complete JSON response). |
| `agent_id` | string | No | ID of an agent to use instead of a base model. Either 'model' or 'agent_id' must be set. |
| `metadata` | object | No | Custom key-value metadata to attach to the conversation for tracking or organization. |
| `description` | string | No | Description of the conversation purpose or context. |
| `instructions` | string | No | System instructions to guide the conversation behavior and style. |
| `agent_version` | string | No | Specific agent version to use. Only applicable when 'agent_id' is specified. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Update Agent

**Slug:** `MISTRAL_AI_UPDATE_AGENT`

Tool to update an existing agent's configuration. Use when you need to modify an agent's name, description, model, instructions, tools, or other settings.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `name` | string | No | Updated name for the agent |
| `model` | string | No | Updated model to use for the agent (e.g., 'mistral-large-latest', 'mistral-small-latest') |
| `tools` | array | No | Updated list of tools available to the model during conversation |
| `agent_id` | string | Yes | The unique identifier of the agent to update |
| `handoffs` | array | No | Updated list of handoff configurations |
| `metadata` | object | No | Updated custom metadata to associate with the agent |
| `description` | string | No | Updated description of the agent's purpose |
| `instructions` | string | No | Updated instruction prompt the model will follow during the conversation |
| `completion_args` | object | No | White-listed arguments from the completion API. |
| `deployment_chat` | boolean | No | Updated flag indicating if agent is deployed for chat |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Update agent version

**Slug:** `MISTRAL_AI_UPDATE_AGENT_VERSION`

Tool to update the current version of an agent (Beta). Use when you need to switch an agent to a different version from its available versions.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `version` | integer | Yes | Version number to set as the current version. Must be a non-negative integer from the agent's available versions list. |
| `agent_id` | string | Yes | Unique identifier of the agent to update (e.g., 'ag_019c5ecef27c7721a09ec60a67fceee5') |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Update library

**Slug:** `MISTRAL_AI_UPDATE_LIBRARY`

Tool to update an existing document library's properties. Use when you need to modify a library's name or description. Note: This is a beta endpoint.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `name` | string | No | New name for the library. If not provided, the current name is retained. |
| `library_id` | string | Yes | Unique identifier (UUID) of the library to update |
| `description` | string | No | New description for the library. If not provided, the current description is retained. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Update library document

**Slug:** `MISTRAL_AI_UPDATE_LIBRARY_DOCUMENT`

Update the metadata of a document in a Mistral AI library. Use this to modify a document's name or add/update custom attributes without re-uploading the file content. This is a beta endpoint.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `name` | string | No | New name for the document. If provided, this will replace the current document name. |
| `attributes` | object | No | Custom attributes to attach to the document as key-value pairs. Can be used to store metadata such as category, priority, tags, or any other relevant information. |
| `library_id` | string | Yes | The unique UUID identifier of the library containing the document. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |
| `document_id` | string | Yes | The unique UUID identifier of the document to update. Must be a valid UUID format (e.g., xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Upload File

**Slug:** `MISTRAL_AI_UPLOAD_FILE`

Upload a file to Mistral AI for use in fine-tuning, batch processing, or OCR. Use this tool to: - Upload .jsonl files for fine-tuning models - Upload files for batch processing requests - Upload documents for OCR processing Maximum file size: 512 MB. For fine-tuning, only .jsonl files are supported.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file` | object | Yes | File to upload to Mistral AI. |
| `purpose` | string ("fine-tune" | "batch" | "ocr") | No | Purpose for the uploaded file. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |

### Upload Library Document

**Slug:** `MISTRAL_AI_UPLOAD_LIBRARY_DOCUMENT`

Upload a document to a Mistral AI library for use with RAG-enabled agents. Use this action to add documents to a library that can be accessed by Mistral AI agents. The uploaded document will be processed and indexed for retrieval-augmented generation. Prerequisites: - First obtain a valid library_id using MISTRAL_AI_LIST_LIBRARIES or MISTRAL_AI_CREATE_LIBRARY. - Supported file formats include PDF, TXT, DOC, DOCX, and other common document types. - Maximum file size is 100 MB per document.

#### Input Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `file` | object | Yes | File to upload. Supported formats include PDF, TXT, DOC, DOCX, and other common document types. |
| `library_id` | string | Yes | Unique identifier (UUID) of the library where the document will be uploaded. Use MISTRAL_AI_LIST_LIBRARIES to get available library IDs. |

#### Output

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `data` | string | Yes | Data from the action execution |
| `error` | string | No | Error if any occurred during the execution of the action |
| `successful` | boolean | Yes | Whether or not the action execution was successful or not |