Skill UI

NVIDIA’s NeMo Guardrails adds runtime safety rails to LLM apps, covering jailbreak detection, hallucination and fact-checking, PII filtering, and toxicity checks via Colang 2.0 rules and integrations for production use on T4 GPUs.

This workflow executes a single-query search using the Perplexity Sonar model, ensuring all returned answers are web-grounded and accompanied by inline citations. It is essential for building high-trust applications, fact-checking tools, or any system requiring verifiable, real-time information retrieval. The process includes API interaction, detailed response parsing (answer, citations, search results), and formatting the output into a usable, source-linked format.

AI-powered search using Perplexity models via LiteLLM and a single OpenRouter API key for real-time, web-grounded answers with citations, ideal for up-to-date research, literature surveys, fact checking, and domain-specific queries beyond model cutoffs.

A domain-agnostic 13-agent pipeline for rigorous academic research on any topic. Eight modes cover full research, quick briefs, paper review, literature review, fact-checking, three-way literature scans, Socratic research dialogue, and systematic review with optional meta-analysis, from question formulation to APA 7.0 reporting.

This skill forces the AI to operate in a non-inferential mode, ensuring every claim—factual, legal, or numerical—is strictly anchored to user-provided materials or verified online sources. It eliminates assumptions, gap-filling, and background knowledge reliance, making it essential for legal research, evidence review, citation checking, and high-stakes document analysis where evidential fidelity is paramount.

Orchestrates a multi-stage fact-checking workflow: capture multi-platform answers, compare knowledge points, perform authoritative verification, and deliver auditable HTML reports with evidence boundaries and per-platform verdicts.