# fairvisor init

- Canonical URL: https://docs.fairvisor.com/docs/cli/init/
- Section: docs
- Last updated: n/a
> Scaffold a policy.json and edge.env.example in the current directory.


`fairvisor init` creates two starter files in the current directory:

- **`policy.json`** — a policy bundle pre-filled for the chosen use-case template
- **`edge.env.example`** — environment variable template for configuring the edge container

## Synopsis

```
fairvisor init [--template=api|llm|webhook]
```

## Options

| Flag | Default | Description |
|---|---|---|
| `--template` | `api` | Starter template to use: `api`, `llm`, or `webhook` |

## Templates

### `api` (default)

Generic API rate limiting by API key. Suitable for REST APIs.

```json
{
  "bundle_version": 1,
  "policies": [{
    "id": "api-rate-limit",
    "spec": {
      "selector": { "pathPrefix": "/", "methods": ["GET", "POST", "PUT", "DELETE"] },
      "mode": "enforce",
      "rules": [{
        "name": "per-key-limit",
        "limit_keys": ["header:x-api-key"],
        "algorithm": "token_bucket",
        "algorithm_config": {
          "tokens_per_second": 100,
          "burst": 200
        }
      }]
    }
  }],
  "kill_switches": []
}
```

### `llm`

LLM token-rate limiting by org claim in JWT. Suitable for `/v1/chat/completions` style endpoints.

```json
{
  "bundle_version": 1,
  "policies": [{
    "id": "llm-rate-limit",
    "spec": {
      "selector": { "pathExact": "/v1/chat/completions", "methods": ["POST"] },
      "mode": "enforce",
      "rules": [{
        "name": "llm-token-budget",
        "limit_keys": ["jwt:org_id"],
        "algorithm": "token_bucket_llm",
        "algorithm_config": {
          "tokens_per_minute": 120000,
          "tokens_per_day": 2000000,
          "burst_tokens": 120000,
          "default_max_completion": 1024
        }
      }]
    }
  }],
  "kill_switches": []
}
```

### `webhook`

Webhook delivery rate limiting by sender IP, suitable for inbound webhook receivers.

```json
{
  "bundle_version": 1,
  "policies": [{
    "id": "webhook-rate-limit",
    "spec": {
      "selector": { "pathPrefix": "/webhooks/", "methods": ["POST"] },
      "mode": "enforce",
      "rules": [{
        "name": "per-ip-limit",
        "limit_keys": ["ip:address"],
        "algorithm": "token_bucket",
        "algorithm_config": {
          "tokens_per_second": 10,
          "burst": 50
        }
      }]
    }
  }],
  "kill_switches": []
}
```

## Generated `edge.env.example`

```bash
FAIRVISOR_EDGE_ID=
FAIRVISOR_EDGE_TOKEN=
FAIRVISOR_SAAS_URL=https://api.fairvisor.com
FAIRVISOR_EDGE_URL=http://localhost:8080
```

Copy this to `.env` or `edge.env` and fill in the values from your SaaS dashboard before running `fairvisor connect`.

## Examples

```bash
# Create default (api) template
fairvisor init

# Create LLM template
fairvisor init --template=llm

# Create webhook template in a specific directory
cd /etc/fairvisor && fairvisor init --template=webhook
```

## Exit codes

| Code | Meaning |
|---|---|
| `0` | Files created successfully |
| `1` | Failed to write files (check permissions) |
| `3` | Unknown template name |