Classifies text for toxicity, severe toxicity, obscenity, threats, insults, and identity-based attacks.
npx skills add https://github.com/DTMC-marketplace/governance --skill detoxify
Please help me install this Agent Skill. Skill: detoxify Source: https://github.com/DTMC-marketplace/governance/tree/main/skills/detoxify Command: npx skills add https://github.com/DTMC-marketplace/governance --skill detoxify
This Skill helps identify and flag toxic comments and harmful content within text, contributing to safer online environments and compliance with content moderation policies.
Use the detoxify skill to analyze the provided text for toxicity.
High-intent search queries and answers about installing and using this skill.
To detect toxic comments, you can use the detoxify model to classify text for various forms of toxicity including severe toxicity, obscenity, threats, insults, and identity-based attacks.
Machine learning models for content moderation can classify harmful content including general toxicity, severe toxicity, obscenity, threats, insults, and identity-based attacks.
Yes, toxicity detection assists in evaluating AI systems against regulatory requirements like Article 9 of the EU AI Act concerning societal risks.
Analyze text for identity-based attacks and severe toxicity by passing the content through the detoxify classification model to receive scores across multiple toxicity categories.
Accurate classification of threats, insults, and obscenity in content moderation requires the Detoxify model for precise detection of harmful user-generated content.
The best way to flag hate speech is to automatically analyze user comments with a toxicity detection model to identify harmful content for review before publication.
wuyoscarAutomate cross-category toxicity benchmarking for Detoxify classifiers.
brainbrewlabsCompile contextual evidence and policy-aligned decision options for flagged content review.
DTMC-marketplaceDetect and filter toxic content in AI inputs and outputs.
ovachieverClassify user input and system outputs across six safety categories.
