Skill UI

This tool conducts a deep integrity audit on experimental results, scripts, and claims generated by AI models. It uses an external, independent reviewer backend to systematically check for common forms of fraudulent reporting, including fake ground truth generation, score normalization fraud, phantom results, and insufficient scope reporting. Use this when rigorous verification of experimental veracity is required before publishing or making claims.

Simulates a multi-round, expert-level patent office examination (Office Action) using advanced AI models. The system rigorously checks patent claims and specifications against global patentability standards (e.g., Novelty, Non-obviousness, Clarity, Enablement), improving IP quality through suggested amendments and comprehensive reports.

This tool provides systematic, cross-model adversarial review for mathematical proofs written in LaTeX. It reads the proof, identifies logical gaps, missing assumptions, and technical errors (e.g., non-uniform convergence, unjustified assertions). The system then fixes these gaps with full derivations, iteratively re-reviews the content until convergence is achieved, and generates a comprehensive audit report. It is essential for academic research and theory paper validation.

This skill provides comprehensive diagnosis and optimization for agent skill libraries. By analyzing historical session transcripts from platforms like Claude Code and Codex, alongside static skill file checks, it scores each skill across 8 dimensions. It identifies issues such as low trigger rates, poor post-invocation user reaction, and incomplete workflow execution, generating a prioritized report (P0/P1/P2) to significantly improve the agent's overall performance and reliability.

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.

Combines a curated Japanese NLP resource dataset with live web research to produce an up-to-date trend report on models, tooling, benchmarks, embeddings, speech, and related subfields.

This skill generates a comprehensive issue report on the current challenges, limitations, and open problems within Japanese Natural Language Processing (NLP) for a specified topic. By combining knowledge from a curated local dataset with real-time web search, it provides a deep dive into ongoing research gaps, evaluating the state-of-the-art and recommending areas for future work.