Skill UI

A deterministic 'Stop hook' designed to enforce session hygiene for AI agents. It checks machine-verifiable facts such as file modification timestamps of learning logs, disk space, and predefined rationalization patterns. This gate ensures that complex tasks lead to captured knowledge, acting as a mechanical safeguard complementing AI self-auditing.

Manages and coordinates complex workflows involving multiple AI agents and tasks. It efficiently distributes work across multiple workers, handles runtime errors, and intelligently merges disparate results into a unified output, making it ideal for large-scale, distributed AI computations.

This guide outlines Dynamic Workflow Mode, enabling coding agents to generate structured, task-local harnesses for complex, multi-step processes. It establishes a disciplined system for temporary workarounds, extracting reusable shared skills, and implementing control pane checkpoints. Use this when tasks require custom loops, multi-agent collaboration, or reliable handoff artifacts.

This guide helps users navigate and understand the complex components of the ECC (Everything Claude Code) ecosystem, including skills, commands, agents, hooks, and rules. It provides structured paths for new user onboarding, feature discovery, installation planning, and troubleshooting, ensuring answers are derived from the current repository state.

This skill provides comprehensive operational controls for long-lived, cloud-hosted agent workloads. It manages the entire lifecycle—from deployment and scaling to observability and security. Features include detailed monitoring (logs, metrics, traces), safety controls (kill switches, least privilege), and robust change management processes (rollout, rollback, auditing). Ideal for mission-critical microservices and background agents.

A formal framework designed for evaluating AI-assisted code generation sessions. It implements Development Oriented Evals (EDD) principles, allowing users to define capability and regression criteria. The system supports multiple evaluation types—code-based, model-based, and human review—and tracks reliability using advanced metrics like pass@k, ensuring robust quality control for AI agents.

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.

A comprehensive guide for optimizing latency-critical systems, such as market data feeds, real-time dashboards, and streaming agents. It details structured methods for mapping 'hot paths,' tracking key metrics (p95, p99 latency, throughput, freshness), and implementing architectural improvements like caching and batching to ensure high performance and data freshness under load.

This skill provides comprehensive security and monitoring for Large Language Model (LLM) powered trading agents. It helps to audit trade decisions, ensure compliance with risk parameters, and mitigate potential risks associated with automated financial operations.

This skill guides the design and rigorous review of goal-oriented agent feedback loops. It helps ensure that complex autonomous multi-step agents are robust, preventing failures such as infinite looping, goal-gaming, or running to completion with incorrect assumptions. It covers defining machine-decidable goals, setting boundaries, and selecting the correct loop type (servo, regulator).

This skill provides access to the Anthropic Claude Managed Agent runtime, which is a cloud-native, persistent environment for running complex, long-running, or asynchronous AI tasks. Unlike local WASM agents, the managed runtime provides a full container, persistent filesystem, and robust state management across multiple turns. It supports the entire agent lifecycle: creation, prompting, status inspection, event logging, and mandatory termination.

A comprehensive, end-to-end methodology designed for software engineers and AI agents to safely and systematically integrate machine learning models into existing non-ML codebases. It guides users through critical phases, including problem framing, data readiness, architectural decoupling (via APIs/services), baseline model implementation, and final MLOps pipeline setup.