Skill UI

Denario orchestrates a multi-agent AI to automate research workflows, covering idea generation, methodology development, computational experiments, and LaTeX publication output.

PufferLib is a high-performance library designed for advanced reinforcement learning simulations and training. It excels in parallel environment simulation, offering optimized vectorization and native support for multi-agent systems. Utilizing algorithms like PuffeRL (optimized PPO+LSTM), it achieves millions of steps per second, significantly accelerating research and deployment across single-agent and multi-agent environments.

Stable Baselines3 is a robust, PyTorch-based library providing production-ready implementations of core reinforcement learning algorithms (PPO, SAC, DQN, etc.). It facilitates the training of single-agent RL models, supports the creation of custom Gymnasium environments, and handles advanced parallel training via vectorized environments. It is ideal for rapid prototyping and rigorous RL experiments.

A powerful tool for architecting production-grade multi-agent workflows. It allows users to define complex, multi-step AI pipelines using structured patterns (sequential, parallel, hierarchical). Key features include defining explicit handoff contracts, implementing robust failure handling, and managing context/cost discipline across long-running processes, ensuring reliable and scalable AI system deployment.

AgentMail provides a comprehensive REST API email infrastructure specifically designed for AI agents. Agents can provision dedicated inboxes, send and receive emails, handle external service verifications (e.g., GitHub signups), and manage real-time inbound webhooks. The system includes a 'karma' mechanism to ensure high domain reputation and prevent abuse, making it reliable for complex automation workflows.

A comprehensive guide covering patterns and architectures for automating code execution using LLMs. It spans simple sequential pipelines (e.g., daily dev scripts) to advanced, RFC-driven multi-agent Directed Acyclic Graph (DAG) orchestration. Ideal for building CI/CD style continuous development pipelines, managing complex state, and automating quality gates without human intervention.

Explains when to call the code-reviewer subagent, how to collect SHAs, fill the template, and act on its feedback so every task, feature, or merge meets quality expectations.

A comprehensive security scanner designed to audit AI agent skills before deployment. It analyzes source code (Python, Bash, JS), detects critical vulnerabilities like command injection, network exfiltration, and unsafe deserialization. Furthermore, it scans markdown files for prompt injection attacks, and assesses dependency supply chain risks, providing a clear PASS/WARN/FAIL verdict and remediation guidance.

This comprehensive guide outlines best practices for building reliable and robust AI agent harnesses. It covers optimizing action spaces, defining structured tools, managing observation formats, and implementing essential error recovery contracts. Use these principles to dramatically improve agent completion rates, manage context budgets effectively, and transition AI prototypes into stable, production-grade autonomous systems.

This protocol defines the structured communication rules for multiple specialized AI agents simulating C-suite roles (e.g., CFO, CTO, CRO). It dictates invocation syntax, response formatting, and enforces strict constraints (such as loop prevention, call depth limits, and isolation rules) to ensure coordinated, non-contradictory, and reliable cross-functional analysis for executive decision-making.

Performs static security analysis on GitHub Actions workflows that incorporate AI coding agents (e.g., Claude, Gemini, OpenAI). This skill helps identify critical data flow vulnerabilities, such as where attacker-controlled input can reach AI agent prompts or internal systems via CI/CD pipelines, going beyond simple expression checks.

This comprehensive guide outlines a rigorous, human-in-the-loop engineering workflow for leveraging advanced AI agents to complete complex implementation tasks. It mandates defining clear completion criteria, decomposing work into manageable units, and strategically routing models (Haiku, Sonnet, Opus) based on task complexity. The workflow emphasizes continuous evaluation, regression testing, security review of invariants and edge cases, and strict cost discipline to ensure robust and reliable software development using AI assistance.