MCP SERVERReach Mistral's chat, Codestral fill-in-the-middle, vision, OCR, Voxtral transcription and Workflow endpoints from a client that is already talking to a different model.
LABSCO SUMMARY
The 8 tools divide into inference — chat, fill-in-the-middle, vision, OCR, transcription — and Workflows, where an execution is started and then polled until it reaches a terminal state. Model choice stays explicit: each tool names the models it accepts, and mistral_vision requires one with vision support. workflow_execute returns RUNNING rather than a result unless it is told to wait, so anything long is asynchronous by default.
An MCP server over Mistral's API, covering both the inference endpoints and Workflow executions
A Mistral account and MISTRAL_API_KEY. OCR and transcription read from a public URL or a file id from the Files API, so local material has to be uploaded first.
One command plus a key — npx mistral-mcp, then supply credentials
USE CASES
Run OCR over the document and take back markdown per page, choosing which pages, whether headers and footers come along, and how tables are rendered.
Transcribe audio from a URL or an uploaded file, hinting the language and asking for diarization and timestamps.
Hand Codestral the code before the cursor and the code after it, and let it fill the middle rather than regenerate the file.
Launch the execution without waiting, poll until it reaches a terminal state, and signal or query it while it is still running.
Route the drafting turn to a Mistral model with your own temperature, seed and response format, and get the token usage back with the text.
Send messages that mix text with image parts to a model that supports vision, and get an ordinary completion back.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
codestral_fimFill-in-the-middle code completion with Codestral.
Given prompt (code preceding the cursor) and suffix (code after the cursor), Codestral writes the middle. Use for editor autocomplete scenarios, code-patching agents, or structured refactors where you know the target boundaries.
Default stop tokens: [] — let the model decide. Override with stop if needed.
mistral_chatGenerate a chat completion using a Mistral model.
When to use:
Returns structured content with the assistant text and token usage. Does NOT stream — use mistral_chat_stream for long outputs with progress updates.
mistral_ocrRun Mistral OCR on a PDF or image, returning structured markdown per page.
Input document is one of:
Options:
pages: array of 0-indexed page numbers or string like "0-5,7".tableFormat: 'markdown' (default) or 'html'.extractHeader / extractFooter: include page header/footer when present.includeImageBase64: embed extracted image bytes as base64 in the response.document_annotation_format: JSON schema for whole-document structured extraction.bbox_annotation_format: JSON schema for extracted image / bbox annotations.confidence_scores_granularity: 'page' or 'word'.Returns pages[].markdown plus optional pages[].hyperlinks, header, footer, images bounding boxes, annotations, confidence scores, and dimensions.
mistral_visionChat completion with multimodal input: text + image_url parts.
Requires a vision-capable model. Accepted:
Each message's content is either a plain string (pure text) or an array of parts { type: 'text', text } / { type: 'image_url', imageUrl }. The image URL can be an https URL or a data: URI base64 payload.
Returns the assistant text + token usage. For non-visual requests, prefer mistral_chat.
voxtral_transcribeTranscribe an audio file to text using Mistral Voxtral.
Accepted models:
Audio source is one of:
Options:
language: ISO-639-1 hint (e.g. 'fr', 'en'). Boosts accuracy when known.temperature: sampling temperature.diarize: return per-speaker segments (default false).timestampGranularities: ['segment'] to return per-segment timestamps.contextBias: list of phrases/terms that should bias the decoder.workflow_executeStart a Mistral Workflow execution.
workflowIdentifier is the workflow name or ID (visible in mistral://workflows). input is a free-form JSON object matching the workflow's input schema.
Modes:
Use deploymentName to target a specific deployment slot when multiple are configured.
workflow_interactSend a signal to or run a query against a running workflow execution.
action=signal: fire-and-forget event; the workflow reacts asynchronously.
name: signal name defined in the workflow.input: optional payload matching the signal's schema.action=query: synchronous read of internal workflow state.
name: query handler name defined in the workflow.input: optional parameters for the query.query_name + result inline.action=update: synchronous request to modify workflow state mid-execution.
name: update handler name defined in the workflow.input: optional payload for the update.update_name + result inline.workflow_statusGet the current state and result of a workflow execution.
Statuses: RUNNING | COMPLETED | FAILED | CANCELED | TERMINATED | CONTINUED_AS_NEW | TIMED_OUT | RETRYING_AFTER_ERROR
Poll until status is COMPLETED (or terminal) when waitForResult was false. result is populated once the workflow reaches a terminal state.
BEFORE YOU INSTALL
8 tools when we connected; the project documents 41. MISTRAL_MCP_PROFILE sets the tool set: core (default) 12, workflows 7, metier-docs 13, admin 41. process_document needs metier-docs; writes such as files_upload and files_delete need admin. The four connectors_* tools may not register for every key.
INSTALLATION
npx mistral-mcp/plugin install mistral-mcp@swih-pluginsclaude mcp add mistral -- npx -y mistral-mcp@latestThis command does not pass MISTRAL_API_KEY: set it in the environment Claude Code starts from. The plugin route above prompts for the key instead.
{
"mcpServers": {
"mistral": {
"command": "npx",
"args": [
"-y",
"mistral-mcp@latest"
],
"env": {
"MISTRAL_API_KEY": "<YOUR_MISTRAL_API_KEY>"
}
}
}
}MCP servers
ericm1018
stabgan
digitalpromptmarket-beep
CodesWhat