> ## Documentation Index
> Fetch the complete documentation index at: https://docs.posthook.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Project Settings

> Configure retries, notifications, and delivery settings

You can configure your project's behavior in the **Settings** tab of the dashboard, or declaratively via [`posthook.toml`](/essentials/config-as-code).

## General

* **Name**: The display name of your project.
* **Domain**: The base domain for your webhook endpoints. Optional for local development projects without a domain. When set, you can use relative paths when scheduling hooks.
* **Custom Authorization Header**: An optional value (e.g., `Bearer my-token`) that Posthook will include in the `Authorization` header of every request.

## Retries

Control how Posthook handles failures (non-200 responses or timeouts).

* **Max Retries**: Number of retry attempts (1–15, plan-dependent).
* **Retry Delay**: Base delay between attempts (5–60 seconds).
* **Strategy**: `fixed` (constant delay) or `exponential` (increasing delay). Available strategies may vary by plan.
* **Backoff Factor**: Multiplier for each retry when using exponential strategy (e.g., `2.0`).
* **Max Delay**: Upper bound on delay between retries when using exponential strategy.
* **Jitter**: Add randomness to retry timing to avoid thundering herd (exponential only).

<Tip>
  Use a shorter delay for time-sensitive hooks, and a longer delay if your endpoint needs time to recover.
</Tip>

**In `posthook.toml`:**

```toml theme={null}
[retry]
attempts = 5
strategy = "exponential"     # "fixed" or "exponential"
delay_seconds = 15
backoff_factor = 2.0         # exponential only
max_delay_seconds = 3600     # exponential only
jitter = true                # exponential only
```

## Notifications

Configure when and how Posthook notifies you about delivery failures.

### Channels

Notifications can be sent via **Email**, **Slack**, or **Custom Webhooks**. All channels receive the same alerts. Available channels may vary by plan.

### Timing

* **Group Wait** (`group_wait_seconds`): When a failure occurs, wait this long before sending a notification so multiple failures get grouped into one alert. Default: `120` seconds.
* **Follow-Up Interval** (`follow_up_interval_seconds`): Minimum time between notifications for the same project. Follow-ups use exponential backoff (e.g., 15m → 30m → 1h → ... → 24h cap). Default: `900` seconds.

Default timing values may vary by plan.

### Anomaly Detection

Posthook tracks historical failure rates per endpoint and compares them against a recent window. If failure rates spike above the baseline, an alert is triggered at the project level.

* **Excess Failure Threshold** (`excess_failure_threshold`): Number of failures above baseline before alerting. `0` means alert on any excess. Default: `0`.
* **Baseline Window** (`baseline_window_mins`): Historical window for calculating the normal failure rate. Default: `60` minutes.
* **Recent Window** (`recent_window_mins`): Window for detecting spikes against the baseline. Default: `5` minutes.
* **Min Deliveries** (`min_deliveries`): Minimum deliveries needed in the baseline window before anomaly detection kicks in. Prevents false alarms on low-traffic endpoints. Default: `5`.

### Recovery

* **Notify on Recovery** (`notify_on_recovery`): Send an alert when a failing endpoint returns to normal. Default: `true`.

### Example config

```toml theme={null}
[notifications]
channels = ["oncall-slack", "eng-email"]
group_wait_seconds = 120
follow_up_interval_seconds = 900
notify_on_recovery = true
excess_failure_threshold = 0
baseline_window_mins = 60
recent_window_mins = 5
min_deliveries = 5
```

## Retention

Control how long Posthook stores terminal hooks (`completed` and `failed`). Pending and in-retry hooks are retained until they reach a terminal state, then the retention clock starts. After the retention period, records are permanently deleted.

* **Maximum**: Determined by your plan (Free: 7, Launch: 30, Growth: 60, Scale: 90 days).
* **Minimum**: 7 days. You can lower retention to any value between 7 days and your plan's maximum.

<Warning>
  Reducing retention permanently deletes records older than the new limit. This cannot be undone.
</Warning>

**In `posthook.toml`:**

```toml theme={null}
[retention]
days = 60
```

## WebSocket Delivery (Beta)

WebSocket delivery allows you to receive webhooks on your local machine without exposing a public IP or using tools like ngrok.

1. Toggle **WebSocket Delivery** on.
2. Run the CLI command provided:
   ```bash theme={null}
   npx posthook listen --api-key $API_KEY --forward http://localhost:3000
   ```
3. Posthook will accept your scheduled hooks and immediately push them down the WebSocket connection to your CLI, which forwards them to your localhost.

<Note>
  This is intended for development only. For production, we recommend using standard HTTPS delivery.
</Note>

## Async Hooks (Beta)

Async hooks let your endpoint return **HTTP 202** to acknowledge receipt, then call back to Posthook when processing completes. This removes the 10-second delivery timeout for long-running tasks.

Toggle **Async Hooks** on in the dashboard to include `Posthook-Ack-URL` and `Posthook-Nack-URL` callback headers with every delivery. This setting is not yet available in `posthook.toml`.

See the [Async Hooks guide](/essentials/async-hooks) for the full implementation walkthrough.

## Applying changes

After editing `posthook.toml`, preview and deploy:

```bash theme={null}
npx posthook diff    # See what will change
npx posthook apply   # Deploy to Posthook
```

See the [Config-as-Code guide](/essentials/config-as-code) for the full workflow.