sbl-probe

sbl-probe

Serial communication and protocol analysis MCP server that gives AI coding assistants direct access to serial ports for reading, writing, decoding, and capturing embedded device output.

Category
Visit Server

README

<!-- mcp-name: io.github.soundbytelabs/probe -->

sbl-probe

Serial communication and protocol analysis MCP server. Gives AI coding assistants direct access to serial ports for reading, writing, decoding, and capturing embedded device output — no more copy-pasting from picocom.

Part of the Sound Byte Labs embedded tooling suite, alongside sbl-debugger for hardware debugging.

Installation

Create a virtual environment and install the package:

python3 -m venv .venv
source .venv/bin/activate

# Install
pip install -e .

# Or with test dependencies
pip install -e ".[dev]"

MCP Configuration

Register the server in your MCP client's config. For most clients, add to .mcp.json in your project root:

{
  "mcpServers": {
    "sbl-probe": {
      "type": "stdio",
      "command": "/absolute/path/to/.venv/bin/python",
      "args": ["-m", "sbl_probe"]
    }
  }
}

Important: Use the absolute path to the Python binary inside your virtual environment. For example: /home/you/sbl-probe-mcp/.venv/bin/python

Restart your MCP client and the tools are available immediately.

Tools

Connection Management

Tool Description
list_ports List available serial ports with USB metadata and by-id paths
open Open a serial connection (port, baud, optional name)
close Close a named connection
connections List active connections with stats

Data I/O

Tool Description
read Read decoded frames using the active decoder (line-oriented by default)
read_raw Read raw bytes in utf8, hex, or base64
write Write data to a connection

Protocol Analysis

Tool Description
set_decoder Change the active decoder on a connection
decode_buffer Run a decoder over a raw data buffer
list_decoders List available decoder names
probe_baud Auto-detect baud rate by scoring printable text across common rates

Capture & Replay

Tool Description
capture_start Start background capture into a ring buffer (with optional filter, trigger, pretrigger)
capture_stop Stop capture, return summary stats
capture_read Query captured frames (filter by regex, time range, last N)
capture_stats Get frame counts grouped by pattern without reading all frames
capture_save Save capture buffer to a JSON Lines file
capture_load Load a previously saved capture

Architecture

sbl_probe/
├── server.py          # FastMCP server, tool wiring
├── transport/
│   ├── base.py        # Transport protocol (structural typing)
│   ├── serial.py      # pyserial wrapper
│   └── manager.py     # Named connection registry
├── decoders/
│   ├── base.py        # Frame dataclass + Decoder protocol
│   └── raw.py         # Line-oriented decoder
├── capture/
│   ├── buffer.py      # Thread-safe ring buffer with query support
│   ├── engine.py      # Background reader thread
│   └── storage.py     # JSON Lines save/load
└── tools/
    ├── connection.py   # list_ports, open, close, connections
    ├── data.py         # read, read_raw, write
    ├── protocol.py     # set_decoder, decode_buffer, list_decoders
    ├── capture.py      # capture_start/stop/read/stats/save/load
    └── diagnostics.py  # probe_baud

Key design decisions:

  • Threaded pyserial with asyncio.to_thread() — keeps the MCP event loop responsive without the complexity of pyserial-asyncio
  • Pluggable decoders — register new decoders by name, swap at runtime via set_decoder
  • Background capture — daemon thread feeds decoded frames into a ring buffer; query with regex, time range, or tail the last N frames
  • Ingress filtering — filter and trigger patterns on capture_start keep the buffer focused on what matters
  • Error dicts, not exceptions — tools return {"error": "..."} instead of crashing the server

Adding a Custom Decoder

from sbl_probe.decoders.base import Frame, Decoder
from sbl_probe.decoders import registry

class MyDecoder:
    @property
    def name(self) -> str:
        return "my_proto"

    def feed(self, data: bytes, timestamp: float) -> list[Frame]:
        # Parse data, return frames
        ...

    def reset(self) -> None:
        ...

registry.register("my_proto", MyDecoder)

Running Tests

pytest                    # 113 tests
pytest -v                 # verbose
pytest tests/test_capture.py  # just capture tests

Dependencies

  • mcp — Official Python MCP SDK (FastMCP)
  • pyserial — Serial port access
  • Python >= 3.11

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured