Skill UI

A comprehensive framework for preparing for high-stakes, adversarial board meetings, investor updates, or fundraising pitches. This guide teaches founders how to master key business metrics (MRR, burn rate, CAC/LTV) cold, anticipate the toughest questions, and structure an honest, confident narrative that maintains credibility even when presenting bad news.

An adversarial thinking partner designed for founders and C-suite executives. It stress-tests business plans, simulates difficult board meetings, and facilitates brutally honest post-mortems, ensuring vulnerabilities are found before competitors or boards do. Use when you need to challenge assumptions and face hard decisions.

Get a deep, multi-round, and adversarial critical review of your research work, ideas, or drafts. This skill simulates the perspective of a senior ML reviewer (e.g., NeurIPS/ICML level) using dedicated external backends (Codex or Manual). It is designed to find logical gaps, missing experiments, and narrative weaknesses, providing brutal honesty to elevate your academic contribution.

Handles IMO/Putnam/USAMO/AIME style problems by generating refined proofs, then running a fresh-context adversarial verifier to catch flaws, finally producing honest confidence reports and verified PDFs when possible.

The Santa Method is a multi-agent adversarial verification framework designed to improve output quality by requiring two independent reviewers (Agent B and Agent C) to pass the content before publication. It eliminates single-agent bias and systematically improves outputs through iterative refinement loops until convergence or manual escalation.

This advanced code review skill forces genuine perspective shifts by employing three hostile reviewer personas: the Saboteur (production breaks), the New Hire (maintainability), and the Security Auditor (OWASP-informed). It ensures that all code reviews are critically rigorous, eliminating superficial "Looks Good To Me" approvals and catching blind spots missed by the original author or general AI models. Use it before merging a PR for maximum confidence.

A comprehensive multi-agent feedback loop inspired by Generative Adversarial Networks (GANs), designed to build complex, production-ready applications. It separates generation from evaluation, using a 'ruthlessly strict' evaluator (QA) to drive continuous iteration. This process ensures high design quality, robust functionality, and originality, overcoming the limitations of single-agent self-critique. Ideal for full-stack feature development and complex product specification implementation.

AgentFlow transforms your existing Kanban board (Asana, GitHub, Linear) into a self-managing, autonomous AI development pipeline. It orchestrates multiple AI agents through defined stages (Backlog to Done), enforcing deterministic quality gates, performing adversarial reviews, and tracking costs—all visible within your existing project management UI. It provides crash-proof execution and full observability for complex software development lifecycles.

This tool audits the integrity of machine learning experiments by checking for common fraud patterns such as fake ground truth, score normalization fraud, and phantom results. It utilizes an external reviewer backend (like Codex or manual review) to perform an adversarial review across all project artifacts, ensuring reported results are trustworthy and methodologically sound.

This skill performs a rigorous, adversarial review of implemented code. It verifies that the task implementation is complete, structurally sound, respects defined boundaries, and aligns perfectly with approved requirements and design specifications. Use it after an implementer finishes work, or before final acceptance, to ensure code quality and architectural integrity.

A comprehensive offensive checklist for assessing the security and robustness of AI and Large Language Model (LLM) applications. It covers advanced adversarial techniques such as prompt injection, jailbreaking, model extraction, data poisoning, and analyzing system vulnerabilities across components. Essential for red-teaming and security assessment.

This skill performs rigorous mathematical proof verification using an adversarial, cross-model review process. It reads a LaTeX proof, systematically identifies logical gaps (e.g., unjustified assertions, quantifier errors) and theoretical flaws (e.g., illegal interchanges, non-uniform convergence). It then iteratively fixes these gaps with full derivations, re-reviews, and generates a comprehensive audit report, ensuring mathematical rigor and completeness. Use this when seeking deep, academic verification of a theory paper.