@neilurk12/pi-tokenrouter
TokenRouter AI model provider extension for pi coding agent
Package details
Install @neilurk12/pi-tokenrouter from npm and Pi will load the resources declared by the package manifest.
$ pi install npm:@neilurk12/pi-tokenrouter- Package
@neilurk12/pi-tokenrouter- Version
0.1.0- Published
- Jun 13, 2026
- Downloads
- not available
- Author
- neilurk12
- License
- MIT
- Types
- extension
- Size
- 3.1 KB
- Dependencies
- 0 dependencies · 1 peer
Pi manifest JSON
{
"extensions": [
"./dist/index.js"
]
}Security note
Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.
README
pi-tokenrouter
TokenRouter AI model provider extension for pi coding agent.
Installation
pi install npm:@neilurk12/pi-tokenrouter
Setup
Get an API key from TokenRouter, then choose one of the following methods:
Environment variable (recommended)
export TOKENROUTER_API_KEY="your-api-key"
Config file
Create ~/.pi/tokenrouter.json:
{
"apiKey": "your-api-key"
}
If both are set, the environment variable takes precedence.
Custom Models
Override or extend the default model list in ~/.pi/tokenrouter.json:
{
"models": [
{ "id": "MiniMax-M3", "name": "MiniMax M3" },
{ "id": "my-custom-model", "name": "Custom Model", "contextWindow": 256000 }
]
}
When models is provided, it replaces the defaults entirely.
Default Models
| Model ID | Name |
|---|---|
| MiniMax-M3 | MiniMax M3 |
Usage
Select a TokenRouter model in pi with /model command.