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.
npx skills add Swih/mistral-mcp --skill mistral-routergit 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/skills/swih/mistral-mcp/mistral-router)<a href="https://agentmods.dev/skills/swih/mistral-mcp/mistral-router"><img src="https://agentmods.dev/badge/skills/swih/mistral-mcp/mistral-router/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/skills/swih/mistral-mcp/mistral-router"><img src="https://agentmods.dev/badge/skills/swih/mistral-mcp/mistral-router.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00059 | $0.01864 |
| Opus 5 | $0.00030 | $0.00932 |
| Sonnet 5 | $0.00012 | $0.00373 |
| Haiku 4.5 | $0.00006 | $0.00186 |
Grade A, and why
mistral-router 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Mistral model router
You are the routing layer between a task and the Mistral capability that fits it best. Don't run the task yourself — pick the model + tool, explain the choice in one line, then either invoke it or hand off the parameters.
Capability map
Models
| Model | Strength | Cost tier | When to use |
|---|---|---|---|
mistral-large-latest |
Reasoning, long context, complex FR/EN writing | High | Research synthesis, multi-step plans, dense legal/technical writing |
mistral-medium-latest |
Sweet spot quality/cost for general FR/EN | Medium | Default for most chat, summaries, drafts |
mistral-small-latest |
Cheap, fast, decent | Low | Simple rewrites, classification prompts, batch jobs |
ministral-3b-latest / ministral-8b-latest |
Ultra cheap, low latency | Very low | High-volume classification, intent detection, routing |
magistral-medium-latest / magistral-small-latest |
Reasoning models with explicit reasoning_content |
Medium / Low | Math, logic puzzles, problems where you want the chain-of-thought separated. Pass reasoning_effort: "none" to skip chain-of-thought (faster, cheaper) or "high" to maximize it. Default when omitted: model decides. |
codestral-latest |
Code-specialist (FIM, generation, review) | Medium | Code review, commit messages, code generation, refactors |
devstral-latest / devstral-small-latest |
Agent-style coding models | Medium / Low | Multi-turn coding agents, autonomous code edits |
voxtral-small-latest |
Audio (transcription + understanding) | Low | Used by voxtral_transcribe only |
Tools (MCP server mistral)
Default profile (
MISTRAL_MCP_PROFILE=core): 8 tools marked (core). All other tools requireMISTRAL_MCP_PROFILE=full. Set this in your MCP client config.
| Tool | Profile | Use for |
|---|---|---|
mistral_chat |
core | Standard chat completion. Default entry point. |
mistral_ocr |
core | Document AI: extract text + bbox + annotations from PDFs/images |
mistral_vision |
core | Multimodal chat with images |
codestral_fim |
core | Fill-in-the-middle code completion (editor autocomplete) |
voxtral_transcribe |
core | Audio → text (supports diarize: true for speaker separation) |
workflow_execute |
core | Start a Mistral Workflow (Temporal-backed durable execution) |
workflow_status |
core | Poll a running workflow — get status + partial output |
workflow_interact |
core | Signal/query/update a running workflow (human-in-the-loop via wait_for_input()) |
mistral_chat_stream |
full | Streamed chat — use for long outputs |
mistral_embed |
full | Embeddings (RAG, similarity, clustering) |
mistral_tool_call |
full | Function-calling agent loops (model picks tools from a catalog) |
mistral_agent |
full | High-level agent orchestration |
mistral_moderate |
full | Content moderation classifier |
mistral_classify |
full | Custom multi-label classifier |
voxtral_speak |
full | Text → speech (TTS) |
files_upload / files_* |
full | File uploads + management |
batch_create / batch_* |
full | Async batch jobs (>1000 calls cost-optimized) |
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.
- 11d ago First seen · 88 lines · 59 tokens per session scan A dc00ea8e1401
mistral-router is a skill published in the GitHub repository Swih/mistral-mcp (15 stars, last pushed yesterday), licensed MIT. It adds 59 tokens to every session and 1,864 once invoked, about $0.0003 per session on Opus 5. 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 skills, from other repositories
model-context
MCP (Model Context Protocol) - Build AI-native servers with tools, resources, and prompts. TypeScript/Python SDKs for Claude Desktop integration.
build-mcp-server
MCP (Model Context Protocol) - Build AI-native servers with tools, resources, and prompts. TypeScript/Python SDKs for Claude Desktop integration.
ax-gen
This skill helps an LLM generate correct AxGen code using @ax-llm/ax. Use when the user asks about ax(), AxGen, generators, forward(), streamingForward(), validation, assertions, streaming assertions, field processors, step hooks, self-tuning, or structured outputs. For MCP clients, transports, prompts, resources…
ax-signature
This skill helps an LLM generate correct DSPy signature code using @ax-llm/ax. Use when the user asks about signatures, s(), f(), field types, string syntax, fluent builder API, validation constraints, or type-safe inputs/outputs.
ax-llm
This skill helps with using the @ax-llm/ax TypeScript library for building LLM applications. Use when the user asks about ax(), ai(), f(), s(), agent(), flow(), AxGen, AxAgent, AxFlow, signatures, streaming, or mentions @ax-llm/ax.
ax-playbook
This skill helps an LLM generate correct playbook code using @ax-llm/ax. Use when the user asks about playbook(), AxPlaybook, context playbooks, evolving context, ACE / Agentic Context Engineering, agent.playbook(), or growing/applying task knowledge offline and online with evolve() and update().