Server data from the Official MCP Registry
Mistral AI MCP server: chat, OCR, Voxtral audio, Codestral FIM, vision, agents, batch.
Mistral AI MCP server: chat, OCR, Voxtral audio, Codestral FIM, vision, agents, batch.
Valid MCP server (4 strong, 2 medium validity signals). 1 known CVE in dependencies (1 critical, 0 high severity) Package registry verified. Imported from the Official MCP Registry.
6 files analyzed · 2 issues found
Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.
This plugin requests these system permissions. Most are normal for its category.
Set these up before or after installing:
Environment variable: MISTRAL_API_KEY
Add this to your MCP configuration file:
{
"mcpServers": {
"io-github-swih-mistral-mcp": {
"env": {
"MISTRAL_API_KEY": "your-mistral-api-key-here"
},
"args": [
"-y",
"mistral-mcp"
],
"command": "npx"
}
}
}From the project's GitHub README.
MCP server exposing Mistral AI capabilities to any MCP client - Claude Code, Cursor, Zed, Windsurf, Claude Desktop.
Version française : README.fr.md
Mistral has strong models for French, code, OCR, moderation, audio, and agent-style workflows, but most MCP-enabled IDEs default to Anthropic or OpenAI. mistral-mcp gives those Mistral capabilities a clean MCP surface so you can route the right subtask to the right model without rebuilding your agent loop.
The goal of this repo is not "yet another thin wrapper". It aims to be a robust, maintainable MCP server with explicit schemas, predictable outputs, transport flexibility, and good test coverage.
v0.4.3)Core generation:
mistral_chatmistral_chat_streammistral_embedmistral_tool_callcodestral_fimVision and audio:
mistral_visionmistral_ocrvoxtral_transcribevoxtral_speakAgents and classifiers:
mistral_agentmistral_moderatemistral_classifyFiles and batch:
files_uploadfiles_listfiles_getfiles_deletefiles_signed_urlbatch_createbatch_listbatch_getbatch_cancelMCP-native utility:
mcp_sample - delegates generation to the client model via MCP samplingmistral://models - accepted aliases and live model catalogmistral://voices - live voice catalog for Voxtral TTSFrench curated prompts:
french_invoice_reminderfrench_meeting_minutesfrench_email_replyfrench_commit_messagefrench_legal_summaryEnglish curated prompt:
codestral_reviewPrompt enum arguments are wrapped with completable(), so MCP clients can call prompt argument completion via completion/complete.
McpServer API with inputSchema, outputSchema, and annotations on every toolstructuredContent plus text fallbackmcp_sampleDefault mode. This is what Claude Code and most local MCP clients use.
node dist/index.js
Enable with --http or MCP_TRANSPORT=http.
MCP_TRANSPORT=http node dist/index.js
Relevant env vars:
MCP_HTTP_HOST - default 127.0.0.1MCP_HTTP_PORT - default 3333MCP_HTTP_PATH - default /mcpMCP_HTTP_TOKEN - optional bearer tokenMCP_HTTP_ALLOWED_ORIGINS - optional comma-separated allow-listMCP_HTTP_STATELESS=1 - stateless session mode/healthz is intentionally public and does not touch the MCP server.
Run from npm:
npx mistral-mcp
Or install globally:
npm install -g mistral-mcp
mistral-mcp
Build from source:
git clone https://github.com/Swih/mistral-mcp.git
cd mistral-mcp
npm install
npm run build
Set your API key:
export MISTRAL_API_KEY=your_key_here
Or use .env at the repo root. Never commit it.
claude mcp add mistral -- node /absolute/path/to/mistral-mcp/dist/index.js
Example prompt:
Use
mistral_ocron this PDF, then runfrench_meeting_minuteson the extracted text.
npm run dev
npm run build
npm run lint
npm test
npm run inspector
The suite currently contains 151 tests across 4 layers:
MISTRAL_API_KEY is setWithout MISTRAL_API_KEY, the local default is 142 passing plus 9 gated live/stdio tests.
mistral-mcp/
|-- src/
| |-- index.ts
| |-- transport.ts
| |-- tools.ts
| |-- tools-fn.ts
| |-- tools-vision.ts
| |-- tools-audio.ts
| |-- tools-agents.ts
| |-- tools-files.ts
| |-- tools-batch.ts
| |-- tools-sampling.ts
| |-- resources.ts
| `-- prompts.ts
|-- test/
|-- examples/
|-- .github/workflows/ci.yml
|-- package.json
`-- tsconfig.test.json
v0.4.3 — shipped. See CHANGELOG.md for the full diff against v0.3.0:
Runnable scripts live in examples/. See examples/README.md.
MIT Copyright Dayan Decamp
Be the first to review this server!
by Modelcontextprotocol · Developer Tools
Read, search, and manipulate Git repositories programmatically
by Toleno · Developer Tools
Toleno Network MCP Server — Manage your Toleno mining account with Claude AI using natural language.
by mcp-marketplace · Developer Tools
Create, build, and publish Python MCP servers to PyPI — conversationally.
by Microsoft · Content & Media
Convert files (PDF, Word, Excel, images, audio) to Markdown for LLM consumption
by mcp-marketplace · Developer Tools
Scaffold, build, and publish TypeScript MCP servers to npm — conversationally
by Taylorwilsdon · Productivity
Control Gmail, Calendar, Docs, Sheets, Drive, and more from your AI