canonical: https://jentic.com/apis/seldon.local/seldon-external-api

# Seldon External API

seldon.local/engine version 0.1. The API exposes 2 endpoints secured with bearer authentication.

## For AI agents

Programmatically predict, sendfeedback. Covers 2 operations with bearer authentication.

## Scope

Does not handle payments, communications, or crm - use for developer tools only.

## Capabilities

- Predict
- SendFeedback
- Integrate Seldon External API into automated workflows
- Query and filter Seldon External API records by parameters
- Monitor Seldon External API operational status and events

## Use cases

### Developer Tools Operations

Use the Seldon External API to perform developer tools operations programmatically. The API provides 2 endpoints covering core functionality including predict, sendfeedback.

Example prompt: Call POST `/seldon/{namespace}/{deployment}/api/v1.0/predictions` to predict

### Data Retrieval and Monitoring

Query Seldon External API resources on a schedule to track changes, generate alerts, or feed downstream dashboards. Agents poll relevant endpoints, compare against previous state, and trigger actions when thresholds are crossed.

Example prompt: Poll the primary Seldon External API endpoint, compare response to last known state, and alert if changed

### AI Agent Integration via Jentic

AI agents discover and call Seldon External API endpoints through Jentic without managing credentials directly. An agent searches for the required operation by intent, receives the matching endpoint schema, and executes the call with Jentic-managed authentication. This eliminates the need to read API documentation or handle bearer tokens manually.

Example prompt: Search Jentic for 'predict', load the operation schema, and execute with Jentic-managed credentials

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/seldon/{namespace}/{deployment}/api/v1.0/predictions` | Predict |
| POST | `/seldon/{namespace}/{deployment}/api/v1.0/feedback` | SendFeedback |

## Key resources

- **Seldon** — Operations for seldon

## Why Jentic

- **Setup:** Wiring the Seldon External API engine by hand means supplying its bearer token, resolving the deployment {host} and {port} into the right base URL, and building each prediction and feedback request yourself. Through Jentic you install once, import Seldon External from the API Directory, store the token once, and your agent calls it.
- **Permission scoping:** Seldon puts the namespace and deployment in the URL path (`/seldon/{namespace}/{deployment}/...`), so a rule can pin your agent to one deployment: it can send predictions and feedback there and nothing else. You choose the operations it may call, so feedback submission is not included unless you add it.
- **Credential handling:** Your Seldon token is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'send a prediction request to a model' or 'submit feedback on a prediction', and Jentic returns the matching Seldon operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Github** — Alternative developer tools API
- **Gitlab** — Alternative developer tools API

## FAQ

### What authentication does the Seldon External API use?

The Seldon External API uses a Bearer token in the Authorization header. Through Jentic, these credentials are stored encrypted in your Jentic One instance and injected at execution time, so raw secrets never enter the agent context.

### Can I predict with the Seldon External API?

Yes. Use the POST `/seldon/{namespace}/{deployment}/api/v1.0/predictions` endpoint. The API returns structured JSON responses that agents can parse and act on directly.

### What are the rate limits for the Seldon External API?

Rate limits are not specified in the OpenAPI spec. Check the vendor documentation for current limits. Through Jentic, rate limiting is handled automatically with retry logic built into the execution layer.

### How do I predict through Jentic?

Install the Jentic SDK with pip install jentic, authenticate through Jentic One, the self-hosted execution layer, then search for 'predict'. Jentic returns the matching Seldon External API operation with its input schema. Load the schema and execute the call - credentials are injected automatically.

### How many endpoints does the Seldon External API have?

The Seldon External API exposes 2 endpoints covering seldon operations.

### Can I limit what my agent is allowed to do with the Seldon External API?

Yes. Because you run Jentic One yourself and write the rules that decide which operations and credentials your agent may use, you control exactly what it can call on the Seldon External API. Seldon places the namespace and deployment in the URL path, so a rule can pin the agent to a single deployment, and you choose whether it may send predictions, submit feedback, or both. If you only allow the predictions operation, the agent cannot post feedback, and it cannot touch any other deployment.