logo

Making content safety radically
more efficient.

An EU-hosted AI API for real-time CSAM detection — built so harmful uploads are caught before they reach your users.

Start free · 100 credits, no card required

Real-time AI detection

Including new and AI-generated content.

Drop-in REST API

Sub-200ms response. JSON in, JSON out.

EU-hosted, GDPR-first

Inference, storage, and audit logs — all in the EU.

Auditable by default

Every score logged, exportable for legal review.

POSTapi.visork.com / v1 / images / analyze
Request
curl -X POST https://api.visork.com/v1/images/analyze \
  -H "Authorization: Bearer $VISORK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "image_url": "https://your-cdn/uploads/4f2a.jpg",
    "modules": ["csam"],
    "threshold": 0.85
  }'
Response · 184 ms
200 OK
{
  "id": "an_2k9d4e1m7q",
  "verdict": block,
  "modules": {
    "csam": {
      "score": 0.974,
      "matched": true
    }
  },
  "audit_log_id": "log_8c1f02b9",
  "region": "eu-central-1"
}
Score 0.974Threshold 0.85
One credit per analysis · No standing minimum
Audit log retained for review

One REST endpoint, every backend stack.
Built so a junior engineer can ship the integration in an afternoon.

Everything between an upload and a verdict

Tunable thresholds.

Set the precision/recall trade-off per module. Audit-ready scores attached to every verdict.
csam0.974
violence0.412
nudity0.318

Threshold UI — additional modules in development

Speaks every backend.

REST in, REST out. Official SDKs (PHP, Node.js, Python, Java) are on the long-term roadmap.

Verdicts, not raw scores.

Block, quarantine, route to review, or allow — your call, your thresholds.
block
quarantine
route to review
allow

Audit logs by default.

Every decision recorded with score, threshold, and module — exportable for compliance.
log_8c1f02b9block0.974
log_8c1f02b8allow0.012
log_8c1f02b7block0.991
log_8c1f02b6review0.624

Webhook notifications.

Push alerts for flagged content the moment it lands. Asynchronous workflows, native.
POST → /webhook/visork
event: analysis.flagged
Your service
200 acknowledged

Built for the platforms that handle user content

Visork is built for the segments where harmful imagery hits hardest — wherever users upload, the same primitives apply. Start free, scale when you're ready.

Use case · 01
Stop harmful imagery before it ever surfaces in a feed.

Score every upload at the edge of your moderation pipeline. Block the obvious, route the unclear, let the rest through.

Social platforms
Use case · 02
Audit every upload, with EU residency built in.

A drop-in API for object storage and upload pipelines. Inference and audit logs stay inside the EU by design.

Hosting & cloud
Use case · 03
Photo screening at sign-up, frictionless for the user.

Hold profile photos at the gate. Sub-second verdicts mean onboarding stays smooth, even at peak traffic.

Dating & discovery
Use case · 04
Moderate at scale without growing the team.

Quarantine harmful uploads automatically and surface only the borderline cases that genuinely need a human.

Forums & communities
Use case · 05
Catch policy violations the moment listings go live.

Run product photos through a single endpoint before they're indexed — protect your marketplace, not your moderators.

E-commerce with UGC
Use case · 06
Quarantine quietly, escalate confidently.

Webhook-driven workflows for asynchronous review. Every action carries an audit log ready for legal review.

File sharing

Got Questions?

If you can't find what you're looking for, get in touch.

Detection & technology

Integration

Compliance & getting started