Tighten agent runbook

This commit is contained in:
hwhung
2026-04-27 12:03:47 +08:00
parent 3782902257
commit 22bd329ee3
+31 -44
View File
@@ -1,77 +1,64 @@
# Agent Instructions
# Agent Runbook
This repository is a standalone Express API server. It is not intended to be added to a parent workspace.
Run this repository as a standalone app. Do not add it to a parent workspace. Do not move it into `artifacts/`. Do not edit source files for setup.
## Run Exactly
## Setup
Run from this repository root:
```bash
pnpm install --ignore-workspace
pnpm dev
```
The server starts with built-in defaults even when `config.json` does not exist:
If plain `pnpm` is unavailable, use `corepack pnpm` for the same commands. Do not edit `package.json`.
The app runs without `config.json`:
- host: `0.0.0.0`
- port: `8080`
- proxy bearer token: `dev-proxy-key`
- bearer token: `dev-proxy-key`
Do not create `config.json` unless internal provider endpoints are available.
## Optional Provider Config
## Fill Provider Config
Do not create `config.json` unless internal provider endpoints/tokens are available.
Only if internal provider endpoints are available:
```bash
cp config.example.json config.json
```
Then fill only these values:
If they are available, create `config.json` in one write with this shape:
```json
{
"server": {
"host": "0.0.0.0",
"port": 8080
},
"proxy": {
"apiKey": "dev-proxy-key"
},
"providers": {
"openai": {
"apiKey": "internal-openai-token",
"baseURL": "https://your-internal-openai-endpoint/v1"
"apiKey": "",
"baseURL": ""
},
"anthropic": {
"apiKey": "internal-anthropic-token",
"baseURL": "https://your-internal-anthropic-endpoint"
"apiKey": "",
"baseURL": ""
}
},
"logging": {
"level": "info",
"pretty": true
}
}
```
Use internal endpoints supplied by the runtime/integration. Do not use public official provider endpoints.
Fill only `providers.openai.apiKey`, `providers.openai.baseURL`, `providers.anthropic.apiKey`, and `providers.anthropic.baseURL`. Use internal endpoints only. Never use public official provider endpoints.
If provider endpoints are not available, leave config absent. The dashboard, health check, and model list still work. Chat completion calls return `503` until the target provider has both `apiKey` and `baseURL`.
## Routes
- `GET /` and `GET /dashboard` - static model dashboard
- `GET /api/healthz` - health check
- `GET /v1/models` - OpenAI-compatible model list
- `POST /v1/chat/completions` - OpenAI-compatible chat proxy
The `/v1/*` routes require:
```text
Authorization: Bearer <proxy.apiKey>
```
## Verification
## Verify
```bash
curl -s http://localhost:8080/dashboard | head
curl -s http://localhost:8080/api/healthz
curl -s http://localhost:8080/dashboard | head
curl -s -H "Authorization: Bearer dev-proxy-key" http://localhost:8080/v1/models
```
Expected chat result without provider config: `503` with a message asking for `providers.<name>.apiKey` and `providers.<name>.baseURL`.
## Production
```bash
pnpm build
pnpm start
```
Expected without provider config: dashboard, health, and model list work; chat completions return `503`.