Skill UI

This Main Agent acts as a central orchestrator for a specialized squad of AI agents, managing complex software development lifecycles. It guides projects through defined phases—from requirements gathering and planning to architecture, implementation, testing, and deployment. It maintains strict context discipline by summarizing agent outputs and referencing full artifacts, ensuring efficient, structured, and non-overwhelming communication to the user.

This skill implements Autonomous Optimization (AO) using Hypothesis Tree Refinement (HTR). It iteratively improves concrete artifacts (like code, models, or data pipelines) by running structured, long-term experiments against a measurable objective. It maintains a persistent hypothesis tree, coordinating subagents to test hypotheses in isolation while using a held-out test evaluator to prevent overfitting and ensure reliable, cumulative research progress. Ideal for complex benchmark tuning, agent refinement, or optimizing entire workflows.

Aria functions as a System Architect, responsible for establishing the definitive structural foundation of the entire system. She designs comprehensive blueprints covering the data model (schema, relationships, constraints), API contract specifications (endpoints, auth, pagination), modular file structure, and selection of core architectural design patterns. Her output ensures that no code is written without a rigorously defined, approved architectural plan, minimizing ambiguity and technical debt.

This skill acts as a goal-integrity governance layer for complex, high-risk, and multi-phase tasks. It ensures the agent remains strictly aligned with the user's original objective by generating formal Goal Contracts, Phase Checks, Deviation Notices, and Final Audits. Essential for robust backend development, API calls, and validation workflows, preventing silent goal drift.

Dep handles the entire journey from local code to production deployment. This skill automates containerization using Docker and docker-compose, establishes robust CI/CD pipelines (linting, testing, security scanning, deployment), manages environment variables and secrets, and generates Infrastructure as Code (Terraform/Pulumi) configurations. It ensures observability and provides rollback procedures.

This skill maps engineering changes (code diffs, architecture designs, IaC) to specific financial compliance controls, such as PCI-DSS v4.0 and MAS TRM. It generates an auditable report detailing compliance gaps, required evidence, and concrete remediation steps, helping teams catch violations before formal audits.

Max is a specialized agent designed for safe, non-behavioral code improvement. It systematically refactors existing, working code by optimizing time complexity (Big-O), abstracting duplicated logic, removing dead code, and enhancing readability. It operates under a strict constraint: no change may alter the original functionality, requiring all refactored code to pass existing test suites flawlessly.

This skill acts as a literature-research assistant, orchestrating a Python CLI to interface with APIs like Semantic Scholar and arXiv. It enables users to search for academic papers by topic, retrieve detailed metadata, perform impact analysis using citation tracking, download PDFs, and extract text content. It streamlines the entire academic research workflow from initial discovery to deep reading.

Quinn specializes in writing and executing comprehensive test suites to rigorously prove system functionality. This role goes beyond simple pass/fail checks, focusing on mapping acceptance criteria, identifying functional gaps, and covering critical edge cases (e.g., nulls, boundary values, error paths). It validates the system's reliability across Unit, Integration, and End-to-End layers.

Rex is a foundational agent designed to transform vague, high-level user intent into precise, unambiguous, and structured technical specifications. It acts as a requirements analyst, asking clarifying questions, identifying edge cases, and defining scope using frameworks like MoSCoW and User Stories. It ensures that all subsequent development stages (planning, architecture, implementation) work from a single, agreed-upon artifact, minimizing guesswork and scope creep.

An advanced ruleset for crafting clear, precise, and formal English technical prose. This guide integrates established literary principles (like those from Strunk & White and Orwell) with modern best practices for AI-generated content. It is ideal for writing research papers, technical documentation, proposals, and release notes, ensuring maximum clarity and professional tone.

This skill acts as a critical quality gate, verifying claims ("done," "shipped," "fixed") made by AI agents against irrefutable ground truth. Instead of relying on the agent's self-report, it uses the DOS kernel to audit commits and check git ancestry, ensuring that claimed effects actually materialized in the code. Ideal for integrating AI workflows with robust CI/CD practices.