Skill UI

An adversarial Quality Assurance (QA) pass designed to deep-dive into any codebase for logic errors, security flaws, and functional bugs. Instead of confirming that code 'looks right,' this skill actively attempts to break the code by simulating hostile inputs, tracing critical data paths, and checking for subtle, real-world correctness issues. Essential for high-stakes, critical code reviews.

This skill implements the LLM Council concept, enabling multiple open-weight language models to collaborate on complex queries. The process follows a three-stage deliberation: 1) Parallel independent responses from selected models; 2) Peer-to-peer ranking of the responses; and 3) Final synthesis by a designated Chairman LLM. Powered by Fireworks AI, it ensures deep, multi-faceted analysis and consensus building for high-stakes decision support.

The Anti-Deception Harness detects manipulative user prompts, such as manufactured urgency, pressure to validate, or undue reliance on authority. It calls an external tool to provide a structured scaffold detailing integrity procedures and counter-evidence, ensuring responses are honest and robust, preventing compromised or sycophantic answers. Ideal for high-stakes decision support.

This advanced skill performs code review by simulating the documented engineering philosophies of historical figures (e.g., Linus Torvalds, Steve Jobs, Ken Thompson). It moves beyond generic AI findings by grounding critiques in specific, citable principles—such as 'good taste' or 'customer experience.' Ideal for high-stakes PRs, architectural reviews, and ensuring deep, pre-submission code hardening.

This skill enforces a rigorous, multi-phase coding workflow (Discuss -> Map -> Decompose -> Execute -> Verify) to ensure code is structurally grounded, complete, and free from invented APIs or assumptions. It is designed for high-stakes, complex, or multi-file engineering tasks, acting as a senior, disciplined coding partner that plans meticulously before writing code.

A structured presentation deck designed for rigorous review of safety incidents, security risks, and policy compliance. It utilizes distinct visual cues, including hazard stripes, L1/L2/L3 severity tiers, and dedicated code blocks for policy review (e.g., YAML). Ideal for red team exercises, post-incident reviews, and high-stakes technical briefings.

Automates the end-to-end process of launching paid ad campaigns across major digital platforms like Google, Meta, LinkedIn, and TikTok. It handles campaign structure, advanced audience targeting, bid strategy configuration, and includes mandatory budget safeguards and compliance checks. Use this tool for complex, high-stakes campaigns requiring rigorous pre-launch approval and deep performance monitoring.

This guide outlines best practices for ensuring rigorous reproducibility of complex interaction claims, especially in game theory and multi-agent systems (MAS). It details required evidence for proofs, experimental setups (agents, opponents, seeds, compute), and consistency checks between the paper's claims and the provided artifacts. Essential for high-stakes academic submissions like AAMAS.

This comprehensive guide provides expert strategies for structuring and auditing the "Related Work" section for top-tier NLP conferences (like ACL). It covers citation hygiene (Anthology vs. arXiv), maintaining temporal freshness against rapidly evolving baselines, and articulating clear technical contrasts to prove the novelty of your research. Essential for high-stakes academic publishing.

A comprehensive framework designed to rigorously test the stability of key macroeconomic findings. This program addresses potential biases arising from specification choices, sample periods (e.g., Great Moderation, ZLB, COVID breaks), and methodological variations (SVAR vs. LP). It ensures that the reported headline quantity remains robust even when challenged by a skeptical academic referee, providing a targeted defense rather than simply listing extra tables. Ideal for high-stakes empirical macro papers.

A comprehensive guide for authors preparing a rebuttal and response letter after a major journal revision (e.g., R&R). It structures the response to referees and editors, emphasizing the policy contribution and addressing every comment point-by-point. The guide stresses the need for evidence-led arguments, strengthening the welfare/policy reading, and ensuring reproducibility standards are met to successfully resubmit high-stakes academic research.

This skill provides a comprehensive, end-to-end workflow guide for academic journal submissions, specifically tailored for Acta Electronica Sinica (AES). It maps the entire publishing journey—from topic selection and research design to submission, review tracking, revision, and final publication—into a structured timeline. It helps authors manage complex decision points, adhere to strict deadlines (like the 6-month response window), and avoid common pitfalls, ensuring a systematic approach to high-stakes scholarly publishing.