content-scanning

What problem does it solve?

Scan user-generated content for policy violations such as hate speech, spam, PII exposure, misinformation, and copyright infringement so unsafe or non-compliant content can be flagged before publishing or escalated for review.

Core Features & Use Cases

• Multi-category detection: Checks for hate speech, violence, adult content, spam/scams, misinformation, PII exposure, and copyright issues.
• Structured results: Produces per-category confidence scores, exact flagged excerpts with locations, and a PASS/REVIEW/REMOVE recommendation using configurable thresholds.
• Operational scenarios: Useful for pre-publication moderation, automated bulk scans, triage of reported posts, and compliance audits across text or mixed-content sources.

Quick Start

Scan the provided user content for hate speech, spam, and PII by running the content-scanning skill with the content or content location as the argument.