Skill UI

A comprehensive framework for creating and executing intelligent, goal-oriented action plans (GOAP). This system analyzes preconditions, optimizes costs, and provides adaptive replanning capabilities. It is designed for complex objectives that require multiple, dependent steps, ensuring robust task completion even when unexpected state changes occur.

This skill runs an automated evolution process designed to systematically optimize the seven core policy surfaces of an AI agent harness (e.g., planner, contextBuilder, reviewer). By mutating these policies and scoring the variants in a sandbox, it identifies crucial performance improvements without requiring expensive foundation model retraining. It is ideal for empirical configuration tuning and establishing optimized baselines.

This tool provides deep inspection and auditing capabilities for GEPA (genetic-evolution prompt-adaptation) genomes by utilizing the internal `@metaharness/darwin/gepa` library. It is crucial for advanced AI development pipelines, allowing users to validate the structural integrity of genomes, render the full system prompt derived from a genome, or classify specific failure modes found in an execution transcript. Ideal for CI/CD gating and debugging complex LLM applications.

This tool executes a GEPA learning cycle, automating the evolution and optimization of policy prompts against a scored task corpus. Instead of relying on manual prompt iteration, it measures performance improvements, identifying the best working 'harness genome.' It is ideal for measuring sustainable improvements in task performance and supports a free, cost-estimating dry-run mode before committing to actual model calls.

This skill executes the metaharness score to provide a comprehensive, multi-dimensional readiness scorecard for a codebase. It assesses critical dimensions like compilation confidence, task coverage, tool safety, and memory usefulness. Use it to evaluate if a repository is prepared for harness adoption before initiating development tools or framework setup.

This tool executes the advanced `metaharness/darwin` security benchmark (ADR-155). It rigorously evaluates the security detection capabilities of AI systems against a controlled ground-truth corpus, providing critical metrics like TPR, FPR, and patch-pass rates. It compares the champion model against multiple baselines, offering empirical data crucial for iterative security improvement and drift detection.

This skill generates an enterprise-grade threat model report using the `harness threat-model` command. It categorizes threats across the MCP surface, providing a detailed report including the overall worst severity (clean, low, medium, high) and specific findings. Ideal for pre-launch security reviews and periodic audits when sharing findings with an InfoSec team.

This library provides comprehensive patterns for building robust Clinical Decision Support Systems (CDSS) integrated into Electronic Medical Records (EMR). It ensures patient safety by implementing critical checks such as drug interaction screening, weight/age/renal-adjusted dose validation, and standardized clinical scoring (e.g., NEWS2). The engine uses pure functions, guaranteeing high testability and reliability for critical medical insights.

This guide details the Hexagonal Architecture (Ports and Adapters) pattern, a robust method for decoupling core business logic from external infrastructure concerns. It teaches how to structure applications to make the domain model independent of databases, UI frameworks, or message queues. Implementing this pattern drastically improves testability, maintainability, and the ability to swap out external services without rewriting business rules.

This skill provides comprehensive guardrails for handling Protected Health Information (PHI) in US healthcare software. It guides development teams on data classification, security best practices, access control, and audit logging to ensure strict adherence to HIPAA regulations. Use it when dealing with patient data storage, transmission, or processing to prevent breaches and maintain compliance.

A sophisticated system designed to automatically select the optimal AI agent and model tier (e.g., Haiku, Sonnet, Opus) for any non-trivial task. It replaces manual decision-making by integrating learned patterns, predictive modeling, and a robust 3-tier routing mechanism. The process is fully auditable, providing a detailed routing rationale to ensure transparency and reliable execution.

This utility detects and classifies various anomalies (such as spikes, flatlines, drift, or pattern breaks) within telemetry data streamed from IoT devices. It utilizes Z-score anomaly detection to provide actionable insights. This is crucial for investigating unusual device metrics, ensuring system health before deploying major firmware updates (canary advancement), or triaging fleet-wide operational alerts.