Score the file before it reaches storage
Standalone moderation APIs analyze content you send them, and the intake, gating, and enforcement stay yours. Filestack runs moderation inside the upload pipeline.
The unsafe file was already stored, and possibly served, before you caught it.
The file is scored before it reaches your storage or your users.
Image moderation API
The sfw task returns a safe-for-work score for an uploaded image. Set a threshold and let a Workflow publish, quarantine, or route the file for review. This single task provides the NSFW image-detection API.
https://cdn.filestackcontent.com/
security=p:POLICY,s:SIGNATURE/
sfw/oxf3OqBOSlVSfJiT9IO4
{ "sfw": true }
POST /your/webhook
{ "event": "fs.workflow",
"video_sfw": { "safe": false } }
Beyond NSFW
Moderation is one of several intelligence and security checks that can gate an upload.
Tag-based policy checks
The tags task labels image content, and you map those labels to your own policy categories. Explicit content detection is the sfw score itself.
Phishing detection
Scan uploaded documents for phishing signals in the same intelligence layer.
The full UGC moderation pipeline
A complete Workflow that triages at scale and routes the gray zone to people.
Automated moderation triages the volume and people decide the edge cases. The Workflow routes an ambiguous score to a review queue.
How it compares
Dedicated moderation vendors go deeper on taxonomy. Filestack wins when moderation is one gate in a file pipeline you already run.
| Filestack | Sightengine | Hive | Rekognition | |
|---|---|---|---|---|
| Runs at upload | Yes | You wire it | You wire it | You wire it |
| File intake / storage / delivery | Included | No | No | No |
| Video support | video_sfw | Yes | Yes | Yes |
| Workflow gating built in | Yes | DIY | DIY | DIY |
| Granular category taxonomy | Focused set | Deep | Deep | Broad |
Credit where due. Sightengine and Hive are dedicated moderation platforms whose granular taxonomies and review tooling go well beyond a safe-for-work score, and fine-grained classification is their lane. Filestack’s edge is placement, as an automatic gate inside the pipeline you already own.
Frequently Asked Questions
What is a content moderation API?
A content moderation API programmatically detects unsafe or policy-violating content, such as NSFW imagery, in user-uploaded files. Filestack applies AI content moderation to images and video at upload time and returns safe-for-work scores your application can act on.
How does NSFW detection work at upload?
The sfw task scores an uploaded image for safe-for-work content before it is stored. Because moderation runs inside the upload pipeline, a Workflow can use the score to publish, flag, or quarantine the file automatically rather than after it is already live.
Can I auto-quarantine flagged files?
Yes. A Filestack Workflow reads the moderation score and routes the file automatically: safe files continue to storage and delivery, flagged files are quarantined or sent to a human review queue.
Does it moderate video?
Yes. The video_sfw task moderates video content. Video moderation runs asynchronously through Workflows and reports its result by webhook, which is where building a point-solution integration yourself hurts most.
What happens to files that fail moderation?
Files that fail are handled by the Workflow you define: commonly quarantined away from public storage and routed to a human review queue for a final decision. Automated moderation triages at scale; humans decide the edge cases.
Screen uploads before they go live
Add NSFW scoring and automatic quarantine to your upload flow with a free API key.