Add any OpenAI-compatible endpoint to
Kilo Code as a custom provider. No
.env file needed — it's all in the settings UI.
Kilo Code is one of the maintained successors to Roo Code, which was archived in May 2026.
Kilo Settings → Providers → Custom provider → + Connect:
| Field | Value |
|---|---|
| Provider ID | apimaster |
| Display name | APIMaster.ai |
| Base URL | https://apimaster.ai/v1 |
| API key | your key |
Leave Headers empty unless your endpoint requires custom auth headers.
- Check the model ids you want (
claude-sonnet-4-6,gpt-5.5, …) - Click Add N model(s)
- Click Submit
The provider appears under Connected providers with a CUSTOM tag.
Open a session → model dropdown below the input → pick a model under your
provider group. The label shows as APIMaster.ai / model-id.
| Symptom | Cause | Fix |
|---|---|---|
401 |
Bad key | Re-copy from console |
| Empty model list | Key lacks permission, or endpoint unreachable | curl /v1/models to confirm |
| Bad URL error | Missing /v1 |
Kilo appends only /chat/completions |
| Wrong model responding | Picked from another provider group | Check the dropdown group label |
- Custom provider connected, Base URL ends in
/v1 - Models added and Submit clicked
- Chat shows
<provider> / <model-id> - Test message returns a reply
Before blaming the agent, confirm the endpoint itself works. This repo ships a zero-dependency checker that tests both protocols and prints the exact base_url to paste into each tool:
python check_endpoint.py --url https://apimaster.ai --key YOUR_API_KEYIt reports which protocol the endpoint speaks, TTFT, the advertised model list, and whether your model id is actually on it.
Configs for Claude Code, Kilo Code, OpenCode, Aider, Codex, Cursor, LiteLLM and LangChain: ai-coding-agent-api-configs
Examples use https://apimaster.ai/v1, an OpenAI-compatible endpoint fronting
multiple model providers with one key. Every config here works with any
OpenAI-compatible endpoint — substitute your own, including a self-hosted
new-api or LiteLLM proxy.
MIT