singhpratech/crimson-crab-mcp-template

A minimal, production-ready Model Context Protocol
(MCP) server, written in Rust, backed by Anthropic's Claude API through the
crimson-crab SDK. It exposes five
tools: ask_claude (send a prompt, get Claude's reply), chat (multi-turn
conversation), count_tokens (price a prompt without running it), list_models
(enumerate the models your API key can use), and get_model (limits and
metadata for one model).
Use it as the reference starting point for building your own Claude-powered MCP
tools in Rust.
Use this template
Click **Use this template** on GitHub, or generate a fresh project with
cargo generate:
You can also just clone it:
Quickstart
The server communicates over stdio, so running it directly just waits for an MCP
client to connect. All logging goes to **stderr** — stdout is reserved for the MCP
protocol. Set RUST_LOG=debug for more verbose logs.
To build an optimized binary you can point a client at:
Wire it into Claude Desktop
Add an entry to your Claude Desktop MCP config
(claude_desktop_config.json), pointing at the built binary and passing your API
key through the environment:
Restart Claude Desktop; the ask_claude, chat, count_tokens,
list_models, and get_model tools will then be available.
The tools
ask_claude
| Parameter | Type | Required | Description |
| --------- | ----------------- | -------- | ---------------------------------- |
| prompt | string | yes | The prompt to send to Claude. |
| system | string | no | Optional system prompt. |
Builds a Claude Messages request (defaulting to the claude-opus-5 model),
calls the API, and returns the concatenated text of Claude's reply.
chat
| Parameter | Type | Required | Description |
| ------------ | ---------- | -------- | -------------------------------------------------------- |
| messages | array | yes | The conversation so far, oldest first: {role, content} where role is an enum constrained to "user" or "assistant". Must end with a user turn. |
| system | string | no | Optional system prompt. |
| model | string | no | Model id (defaults to the ask_claude model). |
| max_tokens | integer | no | Maximum tokens to generate (default 1024). |
Sends the whole message history to Claude and returns the next reply — use it
instead of ask_claude when the caller needs to keep context across turns.
count_tokens
| Parameter | Type | Required | Description |
| --------- | ----------------- | -------- | -------------------------------------------------- |
| prompt | string | yes | The prompt whose token count you want. |
| system | string | no | Optional system prompt to include in the count. |
| model | string | no | Model id to count against (counts are model-specific). |
Returns the number of input tokens the prompt would consume, without running
it — useful for estimating cost before an expensive call.
list_models
| Parameter | Type | Required | Description |