Skill UI

Assesses AI/ML models and datasets for ethical risks, fairness concerns, and bias, generating reports and mitigation guidance via the ai-ethics-validator plugin whenever a user requests an ethics review or bias detection.

Helps implement, audit, and validate JSON-LD schema markup so websites earn rich results, clarify content for AI search, and track structured data issues reported in Search Console before launching updates.

This tool audits the token consumption across all components (agents, skills, rules, MCP servers) within a Claude Code session. It analyzes token overhead, identifies bloat (e.g., overly long descriptions, redundant components), and generates a detailed context budget report. Use it to maintain optimal context window usage, improve performance, and prevent context saturation.

This tool automates the end-to-end workflow for GDPR Data Subject Access Requests (DSAR). It handles identity verification, discovers PII across structured and unstructured data sources using NER and regex, maps data to Article 15 requirements, and generates auditable compliance reports, ensuring adherence to ICO/EDPB guidelines and scaling compliance efforts.

Reads a LaTeX proof, finds gaps via cross-model adversarial review, fixes each gap with full derivations, re-reviews until convergence, and outputs an audit report with proof-obligation accounting.

A systematic auditor designed to evaluate the resilience and quality of complex AI workflows. It scans across five critical dimensions—Prompt Quality, Context Efficiency, Tool Health, Architecture Fitness, and Safety—providing a scored report and prioritized, actionable remediation steps. Use this tool to find hidden flaws and ensure robustness before deploying any AI agent or system.

This tool audits the token consumption across all components (agents, skills, rules, MCP servers) within a Claude Code session. It analyzes token overhead, identifies bloat (e.g., overly long descriptions, redundant components), and generates a detailed context budget report. Use it to maintain optimal context window usage, improve performance, and prevent context saturation.

Guides users through the complete GAIA benchmark submission process, including environment validation, cost estimation, benchmark execution, result packaging, and leaderboard comparison. Ensures integrity via audit reports before submission to the Princeton GAIA leaderboard.

This tool provides a holistic, portfolio-level quality assessment by evaluating multiple content pieces (assets, libraries, or deliverables) simultaneously. Instead of reviewing content one by one, it processes the entire set to generate comprehensive rankings, dimension breakdowns (e.g., brand voice, claim verification), systemic issue reports, and a prioritized revision list. Ideal for content audits, pre-launch checks, and large campaign reviews.

Provides comprehensive guidelines for designing, auditing, and documenting empirical evidence for high-level AI research submissions (e.g., AAAI). It covers essential methodologies including strong baselines, single-factor ablations, robustness testing (shift, seed, prompt), statistical significance, and ensuring perfect alignment with reproducibility checklists and ethical reporting. Use this guide before submission to ensure empirical evidence strongly supports all claimed AI contributions.

This skill helps researchers strengthen AAAI paper submissions by ensuring reproducibility standards are met. It covers experimental traceability, seed reporting, compute disclosure, and dataset licensing. Users learn to map claims to evidence, audit checklists, and prepare artifacts for Phase-1 reviewers to validate rigor across AI scopes.

Provides comprehensive guidelines for designing, auditing, and reporting high-rigor multi-agent experiments, especially in game theory and self-play systems. It emphasizes testing the *interaction* claim—such as convergence to equilibrium, truthful behavior, or emergent cooperation—against held-out opponents, populations, and strategic deviations, moving beyond simple single-agent leaderboard scores.