Zequent

Zequent Client SDK — Connector

client.connector() gives your application access to the platform's system of record: your organization's assets, the payloads mounted on them, and the Skill Registry. A client credential (ZQNT_CLIENT_TOKEN) reaches only its own organization's assets. Schedules, policies, technical configuration and organizations are administration: they are managed in the Admin Console, and a client credential is refused (PERMISSION_DENIED) — see what a client credential may call.

Full method-by-method reference: Connector API Reference.

For Python, see CONNECTOR_PYTHON.md.

Requests share a context

Every Connector request carries an optional ConnectorRequestContext for transaction correlation — you don't need to fill it in for normal use; the SDK generates a transaction ID automatically if you leave it unset.

import com.zqnt.sdk.client.connector.domains.*;

var request = GetAssetBySnRequest.builder()
    .context(ConnectorRequestContext.builder().sn("YOUR_DEVICE_SN").build())
    .build();

client.connector().getAssetBySn(request)
    .thenAccept(response -> System.out.println("Asset: " + response));

Looking up an asset

var request = GetAssetByIdRequest.builder()
    .assetId("550e8400-e29b-41d4-a716-446655440000")
    .build();

client.connector().getAssetById(request)
    .thenAccept(response -> System.out.println(response));

Updating an asset, and listing or updating the payloads mounted on it (camera, gimbal, sensor), follow the same request/response shape — see the reference for the full method list.

Skill Registry

The Skill Registry is the platform's persisted catalog of every command an edge adapter has reported, one entry per (commandId, schemaVersion), each with a lifecycle status. Read it to see which commands exist across your fleet — not only on the assets that are online right now:

client.connector().listSkillContracts(null, null)   // (status, commandId) — either may be null
    .thenAccept(contracts -> contracts.forEach(c ->
            System.out.println(c.getCommandId() + " v" + c.getSchemaVersion() + " " + c.getStatus())));

Changing the registry (observeSkillContract, setSkillContractStatus, setSkillContractPermissions) is refused for a client credential — the registry is platform-wide.

Missions and tasks are gone in 2.0: automated work is built as Applications and Skills — see Applications & Skills. No-fly zones are managed in the Admin Console and apply to every flight the platform commands — see No-fly zones and safe returns.

Checking what an asset supports

client.remoteControl().getCapabilities(sn)
    .thenAccept(snapshot -> snapshot.getCapabilities()
            .forEach(c -> System.out.println(c.getCommandId() + " -> " + c.getState())));

Each entry carries the command id, a description, its target type, and an input schema. Use it to build a UI that only shows commands the asset actually implements. What an asset reports comes from its edge adapter — see Edge SDK — Connector.

Error handling

Connector responses carry a success flag rather than throwing for expected business errors (not found, validation failure). Handle both:

client.connector().getAssetBySn(request)
    .thenAccept(response -> {
        if (!response.isSuccess()) {
            log.warn("Asset lookup failed: {}", response.getError().getErrorMessage());
            return;
        }
        // use response
    })
    .exceptionally(err -> {
        log.error("Connector call failed", err);
        return null;
    });

See also

Was this page helpful?