...
npx skills add https://github.com/DTMC-marketplace/governance --skill perspective-api
Please help me install this Agent Skill. Skill: perspective-api Source: https://github.com/DTMC-marketplace/governance/tree/main/skills/perspective-api Command: npx skills add https://github.com/DTMC-marketplace/governance --skill perspective-api
This Skill helps identify and mitigate harmful content, such as toxicity and bias, within text data, supporting compliance with AI regulations.
Use the perspective-api skill to analyze the text 'This is a great product!' for toxicity.
High-intent search queries and answers about installing and using this skill.
To detect toxicity and bias in text, you can use this Skill to analyze text and score attributes like severe toxicity and identity attacks via the Perspective API. It flags comments violating community guidelines to reduce manual review time.
Toxicity detection scores harmful language attributes, such as identity attacks, to support AI ethics. This Skill helps assess AI systems against EU AI Act Article 9 requirements for fairness and risk mitigation by identifying potential biases in language.
To assess AI systems against EU AI Act Article 9 fairness requirements, use this Skill to identify and mitigate language biases. It provides tools for the assessment, implementation, and monitoring of AI systems to ensure regulatory compliance.
Yes, you can use Perspective API for real-time content evaluation. This Skill integrates with third-party APIs to analyze text data on the fly, scoring attributes like toxicity and severe toxicity to automatically flag harmful comments.
Automated bias assessment in text identifies potential biases in language to support fairness, but it relies on Perspective API scoring attributes like toxicity. It serves as a risk mitigation tool for regulatory compliance rather than a definitive measure of AI fairness.
lucasFR2Automate LLM output evaluation with LLM-as-judge pipelines and bias mitigation.
DTMC-marketplaceAudit AI systems for bias and generate fairness reports against EU AI Act Article 10.
laraxotDetect objects in dynamic web content to automate page specification and routing.
DTMC-marketplaceVerify AI system ethics compliance and fairness claims against EU AI Act regulations.
