Skip to content
Back to search

tinymcp

Let LLMs control embedded devices via the Model Context Protocol.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/golioth--tinymcp/badge.svg)](/listings/golioth--tinymcp)

HTML

<a href="/listings/golioth--tinymcp"><img src="/api/listings/golioth--tinymcp/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

To use `tinymcp`, you need to connect a device to the Golioth platform, then run the `tinymcp` server on your local machine.

Firmware

While existing devices can be used without changing the firmware, if you don't already have a device on Golioth, the [blinky example](./examples/blinky) provides a simple demonstration of how to expose LED control to an LLM via `tinymcp`. It uses [Zephyr RTOS](https://www.zephyrproject.org/), which requires [setting up dependencies](https://docs.zephyrproject.org/latest/develop/getting_started/index.html) before building. After doing so, you can initialize your workspace with one of the following commands.

west init -l .

Or, to use the downstream Nordic [nRF Connect SDK](https://github.com/nrfconnect/sdk-nrf).

west init -l . --mf west-ncs.yml

With your workspace initialized, use `west update` to fetch necessary modules and libraries.

west update --narrow -o=--depth=1

You are now ready to build the firmware and flash your device. See the [blinky documentation](./examples/blinky/README.md) for more information.

Server

In order to run the `tinymcp` MCP server locally, the following environment variables need to be supplied.

TINYMCP_PROJECT=<your-golioth-project-id>
TINYMCP_DEVICE=<your-golioth-device-id>
TINYMCP_API_KEY=<api-key-for-your-golioth-project>

The server can be built into an executable using `go build -o tinymcp ./server` or run directly with `go run ./server`.

If you don't already have a device on Golioth, the [blinky example](./examples/blinky) demonstrates how to expose LED control to an LLM via `tinymcp`. See the [documentation](./examples/blinky/README.md) to build and flash on your device, then expose the registered RPCs in your MCP schema.

There are many [MCP client implementations](https://modelcontextprotocol.io/clients), most of which should be compatible with `tinymcp`. Setup for a few popular options are listed below.

MCP Inspector

[MCP Inspector](https://github.com/modelcontextprotocol/inspector) is a visual testing tool for MCP servers. It allows you to interact with MCP methods directly, rather than going through an LLM. This can be useful when troubleshooting why an LLM may be failing to interact with your device.

The [documentation](https://github.com/modelcontextprotocol/inspector/blob/main/README.md) provides information on how to quickly start the inspector. Once it is running, you can point it to your local `tinymcp` MCP server by entering `http://localhost:8080` as the address and clicking `Connect`. You should be presented with an option to `List Tools`, which should result in all of the RPCs in your schema being exposed.

Claude Code

[Claude Code](https://docs.anthropic.com/en/docs/claude-code/overview) is an agentic coding tool that you interact with in the terminal.

With the `tinymcp` MCP server running locally, it can be [added to Claude Code](https://docs.anthropic.com/en/docs/claude-code/mcp) with a single command.

claude mcp add --transport http tinymcp http://localhost:8080 -H 'Authorization: Bearer none'

> [!NOTE] > The `Authorization` header is supplied in this case to indicate to Claude Code > that it should not attempt to authenticate via the OAuth flow.

Gemini CLI

The [Gemini CLI](https://github.com/google-gemini/gemini-cli) is an open-source AI agent that allows you to access [Gemini](https://gemini.google.com) from the terminal.

The [`.gemini/settings.json`](./.gemini/settings.json) file in this repository will automatically configure the CLI to connect to a locally running `tinymcp` MCP server if the CLI is started in this repository.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 125 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # tinymcp
  2. Let LLMs control embedded devices via the [Model Context
  3. Protocol](https://modelcontextprotocol.io/).
  4. > [!WARNING]
  5. > `tinymcp` is an experimental project. Expect frequent breaking changes.
  6. > Extreme caution should be taken when delegating physical capabilities to AI
  7. > systems.
  8. ## Background
  9. The `tinymcp` project includes an MCP server, which is executed locally, but
  10. communicates with remote devices via the [Golioth management
  11. API](https://docs.golioth.io/reference/management-api/). It uses two existing
  12. device services on the Golioth platform, [LightDB
  13. State](https://docs.golioth.io/application-services/lightdb-state/) and [Remote
  14. Procedure Calls (RPCs)](https://docs.golioth.io/device-management/rpc/), to
  15. expose device functionality via MCP.
  16. RPCs can be exposed as
  17. [tools](https://modelcontextprotocol.io/specification/2025-06-18/server/tools)
  18. by including a schema in on the `mcp/tools/schema` path in LightDB State.
  19. Because LightDB State can be updated by devices or users, RPCs on existing
  20. devices can be exposed without needing to modify the device firmware.
  21. ## Setup
  22. To use `tinymcp`, you need to connect a device to the Golioth platform, then run
  23. the `tinymcp` server on your local machine.
  24. ### Firmware
  25. While existing devices can be used without changing the firmware, if you don't
  26. already have a device on Golioth, the [blinky example](./examples/blinky)
  27. provides a simple demonstration of how to expose LED control to an LLM via
  28. `tinymcp`. It uses [Zephyr RTOS](https://www.zephyrproject.org/), which requires
  29. [setting up
  30. dependencies](https://docs.zephyrproject.org/latest/develop/getting_started/index.html)
  31. before building. After doing so, you can initialize your workspace with one of
  32. the following commands.
  33. ```
  34. west init -l .
  35. ```
  36. Or, to use the downstream Nordic [nRF Connect SDK](https://github.com/nrfconnect/sdk-nrf).
  37. ```
  38. west init -l . --mf west-ncs.yml
  39. ```
  40. With your workspace initialized, use `west update` to fetch necessary modules
  41. and libraries.
  42. ```
  43. west update --narrow -o=--depth=1
  44. ```
  45. You are now ready to build the firmware and flash your device. See the [blinky
  46. documentation](./examples/blinky/README.md) for more information.
  47. ### Server
  48. In order to run the `tinymcp` MCP server locally, the following environment
  49. variables need to be supplied.
  50. ```
  51. TINYMCP_PROJECT=<your-golioth-project-id>
  52. TINYMCP_DEVICE=<your-golioth-device-id>
  53. TINYMCP_API_KEY=<api-key-for-your-golioth-project>
  54. ```
  55. The server can be built into an executable using `go build -o tinymcp ./server`
  56. or run directly with `go run ./server`.
  57. If you don't already have a device on Golioth, the [blinky
  58. example](./examples/blinky) demonstrates how to expose LED control to an LLM via
  59. `tinymcp`. See the [documentation](./examples/blinky/README.md) to build and
  60. flash on your device, then expose the registered RPCs in your MCP schema.
  61. There are many [MCP client
  62. implementations](https://modelcontextprotocol.io/clients), most of which should
  63. be compatible with `tinymcp`. Setup for a few popular options are listed below.
  64. #### MCP Inspector
  65. [MCP Inspector](https://github.com/modelcontextprotocol/inspector) is a visual
  66. testing tool for MCP servers. It allows you to interact with MCP methods
  67. directly, rather than going through an LLM. This can be useful when
  68. troubleshooting why an LLM may be failing to interact with your device.
  69. The
  70. [documentation](https://github.com/modelcontextprotocol/inspector/blob/main/README.md)
  71. provides information on how to quickly start the inspector. Once it is running,
  72. you can point it to your local `tinymcp` MCP server by entering
  73. `http://localhost:8080` as the address and clicking `Connect`. You should be
  74. presented with an option to `List Tools`, which should result in all of the RPCs
  75. in your schema being exposed.
  76. #### Claude Code
  77. [Claude Code](https://docs.anthropic.com/en/docs/claude-code/overview) is an
  78. agentic coding tool that you interact with in the terminal.
  79. With the `tinymcp` MCP server running locally, it can be [added to Claude
  80. Code](https://docs.anthropic.com/en/docs/claude-code/mcp) with a single command.
  81. ```
  82. claude mcp add --transport http tinymcp http://localhost:8080 -H 'Authorization: Bearer none'
  83. ```
  84. > [!NOTE]
  85. > The `Authorization` header is supplied in this case to indicate to Claude Code
  86. > that it should not attempt to authenticate via the OAuth flow.
  87. #### Gemini CLI
  88. The [Gemini CLI](https://github.com/google-gemini/gemini-cli) is an open-source
  89. AI agent that allows you to access [Gemini](https://gemini.google.com) from the
  90. terminal.
  91. The [`.gemini/settings.json`](./.gemini/settings.json) file in this repository
  92. will automatically configure the CLI to connect to a locally running `tinymcp`
  93. MCP server if the CLI is started in this repository.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.