Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/Swih/mistral-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/plugins/swih/mistral-mcp/claude-plugin)<a href="https://agentmods.dev/plugins/swih/mistral-mcp/claude-plugin"><img src="https://agentmods.dev/badge/plugins/swih/mistral-mcp/claude-plugin/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/plugins/swih/mistral-mcp/claude-plugin"><img src="https://agentmods.dev/badge/plugins/swih/mistral-mcp/claude-plugin.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
mistral-mcp scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 19d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
{
"$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
"name": "mistral-mcp",
"version": "1.0.0",
"description": "Extract invoices and documents from text, Markdown or Mistral OCR into typed JSON. Six core MCP tools, chat, vision, transcription and 11 Claude Code skills.",
"author": {
"name": "Dayan Decamp",
"email": "[email protected]",
"url": "https://github.com/Swih"
},
"homepage": "https://github.com/Swih/mistral-mcp",
"repository": "https://github.com/Swih/mistral-mcp",
"license": "MIT",
"keywords": [
"mistral",
"mcp",
"document-extraction",
"invoice-extraction",
"structured-output",
"ocr",
"codestral",
"voxtral",
"french",
"ai",
"vision",
"audio"
],
"userConfig": {
"mistral_api_key": {
"type": "string",
"title": "Mistral API key",
"description": "Your Mistral API key from https://console.mistral.ai/. Configure it using your client's supported secret settings; access and quota depend on your account.",
"sensitive": true,
"required": true
}
}
}
What it installs
The manifest is a name and a version. 11 skills travel with it, and installing the plugin installs all of them — 440 tokens a session between them. Each is measured on its own page, and each can be installed alone.
- Skill mistral-router A 49 tokens
- Skill contract-analyzer A 38 tokens
- Skill pdf-invoice-extractor A 39 tokens
- Skill audio-dispatch A 37 tokens
- Skill french-meeting-minutes A 36 tokens
- Skill french-commit-message A 46 tokens
- Skill compliance-audit-workflow A 38 tokens
- Skill contract-review-workflow A 37 tokens
- Skill research-pipeline-workflow A 38 tokens
- Skill french-invoice-reminder A 42 tokens
- Skill codestral-review A 40 tokens
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 19d ago Changed · +3 lines e3101b9edc01
- 1mo ago First seen · 35 lines scan A 809f85c4cdc2
mistral-mcp is a plugin published in the GitHub repository Swih/mistral-mcp (15 stars, last pushed 18d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other plugins, from other repositories
claude-prompts
Model Context Protocol (MCP) server for versioned prompt templates, multi-step workflow chains, and quality gates. Symbolic syntax: >>prompt --> >>chain @framework :: 'gate'.
skmtc
Author Skmtc generators — derive application code from an OpenAPI document treated as a domain model.
prompt-pilot
Turns rough, vague prompts into precise, context-enriched prompts. Explores the current codebase read-only, weaves real file paths and symbols into the rewritten prompt, and never implements the task itself.
shipshitdev-ai-agents
AI agent development and prompt engineering.
llm-testing-expert
Evaluation strategies, prompt engineering, red teaming, and regression testing for LLMs.
prompt-engineering-expert
Designing effective prompts: structure, few-shot, chain-of-thought, output formatting, and evaluation.