Developer Documentation

API Endpoints

Integrate our state-of-the-art neural forensic engine into your workflow. High-performance, low-latency detection for images and video streams.

POST

/api/detect

Analyze an image or video file for manipulation. This endpoint uses a multi-layered forensic model to return a probability score and specific heatmaps of detected anomalies.

Request Parameters

media_url Required

Public URL of the media file to analyze (JPG, PNG, MP4).

sensitivity Optional

Float between 0.1 and 1.0. Default is 0.5.

webhook_url Optional

Asynchronous result notification endpoint.

cURL Example content_copy

curl -X POST "https://api.obsidian-security.ai/v1/detect" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "media_url": "https://assets.cdn.com/v/042.mp4",
    "sensitivity": 0.8
  }'

info

AI Insight: Batch Processing

For volumes exceeding 5,000 requests per hour, please utilize our Batch Architecture to ensure priority queue placement and reduced overhead.

GET

/api/stats

Retrieve your account's real-time detection metrics, quota usage, and historical accuracy performance.

Response Structure

The response returns a JSON object containing global counters and time-series data for the last 24 hours. Perfect for building internal monitoring dashboards.

speed 200ms Latency

lock SSL Required

JSON Response Status: 200 OK

{
  "quota_used": 14205,
  "quota_limit": 50000,
  "avg_confidence": 0.982,
  "flagged_threats": 12,
  "uptime": "99.998%",
  "request_id": "obs_6741x0"
}

Ready to test?

Our Interactive Playground allows you to run live queries against our sandbox environment before writing a single line of production code.