Starting the Prism Plugin on Your Node

Starting the Prism Plugin on Your Node
Photo by Midhun Harikumar / Unsplash

Lightning Prisms have been implemented as a CLN plugin.

To learn more about CLN's plugin architecture, check out this post.

We are going to jump right into how you can run ROYGBIV on your own CLN node (assuming that you already have a node up and running). This post will be geared towards the Bolt 12 Prism Plugin, but can be used generally for other plugins as well.


ROYGBIV is written in Python with the pyln-client library, so your node will need to have Python installed as well as the pyln-client library. You'll also need the plugin itself.

If you are using the ROYGBIV Stack, then everything is all set and you can just refer to the docs there.


The plugin can be found here. You just need ``, but feel free to clone the whole repo as well.


Python just needs to be installed in the same environment as your node. This means that if you are running CLN on a server, simply install Python onto the server. Or, for example, if you are running CLN in a Docker container, make sure Python is installed there.

apt install python3


Once Python is installed, installing the library is as simple as:

apt install pip && pip install pyln-client

Okay, now that we've got the dependencies taken care of lets jump into actually running the plugin on your node.

For reference, you can check out the Dockerfile we use in the ROYGBIV Stack or the CLN docs.

Starting the Plugin

There are two options to start the plugin.

  1. Generally, you will want the plugin to start whenever your node starts
  2. Often for development, it's handy to manually stop and start the plugin

Auto Start

To make the plugin start when your node starts, it's as simple as adding the path to the plugin in your config file.


Add to your node's config file:

BOLT12 is currently experimental and is subject to change. Make sure your node can create offers.

You can also add the plugin and enable offers in your startup command

Start Up Command

From the command line, when you start your node add the --plugin flag. It would look something like this:

lightningd --plugin=/path/to/ --experimental-offers

Awesome, now your node is running and has exposed the RPC methods made available by the plugin.

But what if you are developing on the plugin? Well, if you make any changes to the plugin file, then those changes will not be reflected until you stop and start (aka restart) the plugin.

Manual Start

In order to stop and start the plugin you will need access to lightning-cli. If you've followed the install instructions for CLN you should be all good to go!

To verify try running lightning-cli getinfo . This should return a JSON object (not an error 😁).

Sometimes (if not specified in your config file), you will need to specify which directory your node is running from using the --lightning-dir flag.

This would look something like this:

lightning-cli --lightning-dir=/path/to/lightning

Start Plugin

lightning-cli plugin start --plugin=/path/to/

Stop Plugin

lightning-cli plugin stop

Verify the plugin is active by running lightning-cli plugin list

Again, this method of manually stopping and starting is generally for development purposes because the plugin needs to be restarted before any changes made are seen by the node.

Tip: in the directory of the plugin use ls | enter -s '' to auto restart whenever changes are made to your plugin file. See here.


Refer to plugin docs, especially the README in the testing directory.



developing = building + learning