canonical: https://jentic.com/apis/parkingaid.io/parkingaid

# ParkingAid Prediction API

ParkingAid Prediction API provides minute-by-minute parking availability forecasts for specified locations using machine learning models trained on historical occupancy data. Given a latitude, longitude, search radius, and forecast timeframe, the API returns availability scores with optional confidence metrics for nearby parking blocks. Ideal for apps that help drivers find parking before arriving at a destination.

## For AI agents

Get minute-by-minute parking availability predictions for locations based on historical data and ML forecasts.

## Scope

Does not reserve parking spots, process payments, or control parking hardware - use only for obtaining availability forecasts based on historical and real-time data.

## Capabilities

- Retrieve parking availability predictions for a geographic location with configurable radius
- Specify forecast timeframes such as 2h, 4h, or custom durations
- Receive minute-by-minute availability scores for nearby parking blocks
- Optionally include confidence metrics for each prediction to assess reliability

## Use cases

### Pre-Trip Parking Planning

Integrate ParkingAid predictions into navigation or ride-hailing apps to help users plan arrivals around parking availability. POST `/v1/predictions` with the destination coordinates, radius, and departure time offset (as timeframe) returns a forecast of availability scores. Users can adjust departure times or choose alternate locations based on predicted ease of parking.

Example prompt: POST `/v1/predictions` with location (lat, lng, radius), timeframe (e.g., '2h'), and include_confidence=true. Parse the blocks array and present the availability_score for each block in the response.

### Event Parking Guidance

For event venues or stadiums, provide attendees with predicted parking availability windows. Query the API before event start time to surface high-confidence blocks with favorable scores, reducing circling and congestion. Embed predictions in event confirmation emails or venue apps.

Example prompt: POST `/v1/predictions` with the venue coordinates, a 500m radius, and a 3h timeframe covering the event window. Filter blocks by availability_score > 0.7 and confidence > 0.8 for reliable guidance.

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/v1/predictions` | Retrieve parking availability predictions for a location |

## Key resources

- **Predictions** — Minute-by-minute parking availability forecasts for blocks near a location
- **Blocks** — Individual parking areas with location, availability score, and optional confidence metric

## Why Jentic

- **Setup:** Wiring the ParkingAid Prediction API by hand means setting up its bearer auth, shaping the `/v1/predictions` request, and handling retries yourself. Through Jentic you install once, import ParkingAid from the API Directory, store the token once, and your agent calls it.
- **Permission scoping:** ParkingAid carries the location and time window in the request body, so scoping is by operation: limit the agent to the operations it needs, such as requesting availability forecasts. Since this API only returns predictions, the agent stays limited to read-only forecasting.
- **Credential handling:** Your ParkingAid bearer token is stored once, encrypted, by your own Jentic One instance and injected as the Authorization header at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'parking availability forecast' or 'predict parking near a location', and Jentic returns the ParkingAid predictions operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **ParkMobile API** — ParkMobile handles parking session payments while ParkingAid predicts where parking will be available.
- **ParkWhiz API** — ParkWhiz provides parking search and booking while ParkingAid forecasts on-street availability.

## FAQ

### What authentication does the ParkingAid API use?

The API uses bearer token authentication. Jentic stores the bearer token encrypted and injects it as the Authorization header at execution time.

### How accurate are the parking predictions?

Accuracy depends on the historical data density for the queried location and the confidence score returned per block. Higher confidence scores indicate more reliable predictions. Include confidence=true in the request to assess reliability.

### What units are used for radius and timeframe?

Radius is specified in meters. Timeframe is a duration string like '2h' (2 hours) or '30m' (30 minutes). The API returns predictions covering the specified duration from the current time.

### Can I limit what my agent is allowed to do with the ParkingAid Prediction API?

Yes. Because you run Jentic One yourself, your own rules decide which ParkingAid operations and credentials the agent may use. This API exposes a single read-only operation, the parking availability forecast call to POST `/v1/predictions`, so you can scope the agent to just that request and nothing more. The agent stays limited to requesting forecasts and cannot reserve spots, take payments, or reach any other endpoint.