Vision AI Agents — Authentication

Vision AI Agents uses API key authentication to secure all platform requests.

Every request made to the Vision AI Agents API must include a valid API key in the request header.

API keys identify the developer application and enforce usage limits, rate limits, and access permissions.


Authentication Method

Method Description Required
API Key Requests must include an API key in the Authorization header Yes
Bearer Token API keys are sent using the Bearer authentication scheme Yes

API Key Authentication

Authentication is performed using a bearer token in the HTTP Authorization header.

Example header:

Authorization: Bearer YOUR_API_KEY

Your API key must be included in every request sent to the platform.

If a request does not include a valid API key, the API will return an unauthorized response.


Authentication Header Format

Header Value Format Description
Authorization Bearer YOUR_API_KEY Authenticates the request using the developer API key
Content-Type application/json Specifies JSON request format

Example Authenticated Request

The following example demonstrates a request with authentication included.

POST /api/video/ingest
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "video_url": "https://example.com/video.mp4"
}

The API validates the API key before processing the request.


Generating API Keys

API keys can be generated from the Vision AI Agents developer dashboard.

Developers can create multiple keys for different applications or environments.


API Key Environments

Environment Purpose
Development Used for testing integrations during development
Staging Used for pre-production validation
Production Used for live applications and production workloads

Using separate keys allows developers to isolate environments and control usage independently.


API Key Permissions

API keys inherit permissions from the developer account that created them.

Permission Scope Description
Video Ingest Upload and register videos for analysis
Video Intelligence Run analytics and intelligence extraction
Audience Testing Request audience engagement analysis
Search Query indexed video intelligence signals

Enterprise accounts may receive additional permissions or expanded usage limits.


Securing API Keys

API keys should always be treated as sensitive credentials.

Best Practice Description
Do not expose keys Never embed API keys in client-side code
Use server environments Store API keys securely in backend infrastructure
Rotate keys regularly Replace API keys periodically to reduce risk
Restrict access Limit API key usage to trusted systems

If an API key is exposed or compromised, it should be revoked immediately and replaced with a new key.


Unauthorized Requests

If a request is sent without a valid API key, the API returns an HTTP 401 Unauthorized response.

Example response:

{
  "error": "unauthorized",
  "message": "Invalid or missing API key"
}

Unauthorized Response Fields

Field Type Description
error string Machine-readable error code
message string Human-readable error description

Developers should verify that the Authorization header is correctly formatted and that the API key is valid.


Rotating API Keys

API keys should be rotated periodically to maintain security.

Recommended rotation workflow:

  1. Create a new API key in the developer dashboard
  2. Update application configuration
  3. Confirm requests succeed using the new key
  4. Revoke the previous key

This ensures uninterrupted service while maintaining security best practices.


API Key Limits

API keys inherit the rate limits defined by the developer usage tier.

Limit Type Description
Concurrent Requests Maximum number of simultaneous API requests
Daily Requests Total API requests allowed per day
Monthly Token Usage Total processing tokens available per billing cycle

Developers can review usage limits in the Rate Limits & Usage Tiers documentation.


Related Documentation

Developers integrating Vision AI Agents should also review:

  • Getting Started
  • API Reference
  • Platform Architecture
  • Rate Limits & Usage Tiers
  • Error Handling