Configure content safety guardrails for AI Gateway endpoints.
npx skills add https://github.com/truefoundry/tfy-agent-skills --skill guardrails-truefoundry
Please help me install this Agent Skill. Skill: guardrails Source: https://github.com/truefoundry/tfy-agent-skills/tree/main/skills/guardrails Command: npx skills add https://github.com/truefoundry/tfy-agent-skills --skill guardrails-truefoundry
This Skill helps you implement robust content safety and security measures for your AI applications, preventing harmful outputs and protecting sensitive data.
Use the guardrails skill to create a new guardrail config group named 'my-safety-group' with PII and content moderation integrations.
High-intent search queries and answers about installing and using this skill.
To configure content moderation guardrails for an AI Gateway, you create a guardrail config group and define rules that apply safety services like PII filtering and prompt injection detection to specific models, users, or MCP tool integrations.
PII filtering and prompt injection detection in LLM security are safety mechanisms that scan user inputs and AI responses to remove sensitive personal data and block malicious prompts attempting to manipulate model behavior.
Yes, you can apply custom guardrail rules to specific MCP tool integrations by defining policies that specify exactly which guardrails apply to which models, users, or tools within your AI Gateway configuration.
Yes, the guardrails configuration supports multiple safety providers at once, allowing you to integrate various services like PII filtering, content moderation, and prompt injection detection simultaneously within a single config group.
You need content safety guardrails for your customer-facing chatbot to prevent harmful outputs, protect sensitive user data from PII exposure, block prompt injection attacks, and ensure compliance with a safe user experience.
passionseedConfigure and enforce content safety guardrails for TrueFoundry AI Gateway.
IgorGanapolskyBlock unsafe or non-compliant crypto trades before execution.
m13vFilter user-generated content and model responses with Google's Model Armor templates.
CaseMarkDetects and redacts PII in PDFs via regex and AI semantic analysis.
