Skill UI

Invariant-Guard enforces a correctness-first approach to algorithm design. Instead of writing code and hoping it works, this skill mandates defining the function contract, loop invariants, termination arguments, and all potential edge cases *before* implementation. It is essential for complex algorithms where subtle bugs (like off-by-one errors or missed post-conditions) are common, ensuring mathematically rigorous and verifiable code.

A powerful tool for designing and writing deterministic, multi-agent workflow scripts in JavaScript. It allows users to orchestrate complex, repeatable, multi-step tasks using advanced patterns like fan-out, pipelines, loops, and judge-panels. Ideal for automating sophisticated business processes that require multiple specialized sub-agents and robust context management.

This skill provides a direct shell call to capture the current Git diff (staged, working-tree, or specific file list) as raw text. It is designed for workflows requiring repository diffs without the overhead of an LLM agent loop, offering significant performance improvements over traditional sub-agent calls.

A lightweight utility designed for fetching content from a single URL via HTTP or HTTPS. It supports various methods (GET, POST, PUT, DELETE) and allows configuration of timeouts and maximum response body sizes. This tool is ideal for quick, simple data extraction steps within a workflow DAG, offering significant performance gains over complex LLM agent loops. Note that complex tasks like JS rendering or multi-page crawling require dedicated sub-agents.

An open-source automation framework for conducting multi-turn, adversarial red-teaming against conversational LLMs. PyRIT simulates real-world attacker behavior using advanced orchestrators (Crescendo, TAP) and scorer feedback loops. It is designed for comprehensive security assessments of AI agents and chatbots, going far beyond single-shot prompt testing to evaluate robustness under gradual or adaptive attacks.

This library provides a comprehensive framework for managing bounded AI workflows (loops). Users can analyze existing codebases or work histories to discover opportunities for repeating, automated agent processes. It guides the user through finding published templates, auditing and repairing weak agents, adapting parameters, or crafting entirely new, robust automation designs, ensuring clear triggers, verification, and mandatory stopping conditions.

WGM is a robust, agentic methodology designed to transform vague or rough requirements into working, validated software. It enforces a disciplined workflow across six stages: deep alignment (Grill), structured planning, deterministic iterative building (Loop), mandatory holdout-scenario judging, and final documentation audits. It ensures that the resulting code converges against rigorously tested acceptance criteria rather than relying on single-pass generation.

This pattern is designed to solve context limitations in complex multi-agent workflows. Instead of relying on a single, massive context dump, it implements a structured, multi-cycle loop (Dispatch -> Evaluate -> Refine -> Loop). This process iteratively refines the search query by scoring content relevance and identifying missing context gaps, ensuring sub-agents gather the most targeted and highly relevant information needed for tasks like debugging or feature implementation.

Orchestrates bootstrapping a working MVP from a design or spec document: ingest the SDD/PRD, plan thin vertical slices, scaffold the first end-to-end slice, then drive a generator-evaluator build loop with review gates and gated feat commits.

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.

A comprehensive, multi-stage skill designed to simulate a professional Continuous Integration (CI) pipeline. It guides users through critical quality gates, including build verification, type checking, linting, running test suites (with coverage checks), security scanning for secrets, and final diff review. Use this skill before creating a Pull Request or finalizing any significant feature to ensure code robustness and adherence to best practices.

Provides comprehensive guidelines for designing rigorous technical evaluations for interactive systems, specifically tailored for top-tier venues like UIST. It emphasizes matching evaluation methods to specific artifact claims, reporting distributions over means, closing the interaction loop, and performing thorough ablations against real-world baselines. This framework ensures the scientific rigor and credibility of submitted research.