CLIProxyAPI is a tool that bridges the gap between command-line interfaces and AI models, enabling seamless integration and interaction.
- https://help.router-for.me/
- https://github.com/router-for-me/CLIProxyAPI
- https://github.com/router-for-me/CLIProxyAPIPlus
- Installer project : https://github.com/brokechubb/cliproxyapi-installer
- install
./aistack cpa install
- launch
./aistack cpa launch
-
authenticate to gemini cli
- note : The local OAuth callback for Gemini via OAuth uses port 8085.
./aistack cpa login gemini-oauth [--project_id <your_project_id>]- if you are on a Google One plan, select it. It is the free tier with google personnal account
- alternative : go to http://localhost:8317/management.html go to OAuth login / Gemini CLI OAuth
-
info
./aistack cpa info
- sample test of launched cpa : list of existing endpoints
curl -s http://localhost:8317
- list available models
YOUR_TOKEN="xxxx"
curl -X GET "http://localhost:8317/v1/models" \
-H "Authorization: Bearer $YOUR_TOKEN" \
-H "Content-Type: application/json"
- OR
./aistack cpa model list
- test a chat completion
curl -X POST http://localhost:8317/v1/chat/completions \
-H "Authorization: Bearer $YOUR_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-2.5-pro",
"messages": [
{"role": "user", "content": "Hello from test"}
]
}'
- enable/disable usage statistics tracking - see it in management console
./aistack cpa set "usage
-statistics-enabled" "true"
./aistack cpa set "usage
-statistics-enabled" "false"
You will have to use the web browser of your current host to auth to gemini. And launch a local port forwarding with ssh to establish a tunnel
On your current host, where run your vscode desktop, use a ssh client and launch ssh -L 127.0.0.1:8085:127.0.0.1:8085 root@xxx.xxx.xxx.xxx -p 22.
- Launch
ssh -L 127.0.0.1:8085:127.0.0.1:8085 remote_ssh_user@remote_ssh_host -p remote_ssh_port - Then
./aistack cpa login gemini-oauth - Then open the browser by following the link to auth
Instead ssh command you can use the "forward port" functionnality in vscode :
- Ports/Add port :
- Port : localhost:8085
- Change Local Adress Port : 8085
- remote_ssh_host:remote_ssh_port is the SSH address you are currently connected through VS Code remote SSH
About SSH port forwarding : https://gist.github.com/StudioEtrange/d10c0b4f17a60b219e0b5722968d5b8c
- In KiloCode extension settings
- API Provider : OpenAI Compatible
- Base URL : http://127.0.0.1:8317/v1
- API Key : <YOUR_TOKEN>
- Model : Kilo Code will load list of available modeles (ex:gemini-2.5-pro)
- Context Window Size : Get size by selecting API Provider of the proxified LLM (ie: Google Gemini default setup for gemini-2.5-pro is 1048576)