Skill UI

This guide provides a systematic framework for authorized email security reviews. It covers advanced phishing analysis, including deep dives into email headers, assessing SPF/DKIM/DMARC compliance, identifying Business Email Compromise (BEC) patterns, and researching mailbox token abuse. It is essential for incident response, threat intelligence, and security architecture review.

This skill implements a comprehensive logging guard to prevent Protected Health Information (PHI) leakage into logs, traces, and error reports. It acts as a pre-emission filter, redacting structured identifiers (SSN, email, phone) via regex and handling free-text PHI using advanced models. Essential for maintaining strict data privacy and compliance when shipping logs to centralized stores or error trackers.

This skill performs the crucial ETL process of mapping OpenMed-extracted, terminology-coded clinical facts (conditions, drugs, measurements) into the OMOP Common Data Model (CDM) v5.4 tables. It is used downstream after entity extraction and source coding, transforming raw NLP findings into standardized, OMOP-conformant records for OHDSI analytics.

This tool systematically audits and scores defined AI agents to ensure they are correctly scoped, highly specific, and safe for programmatic orchestration. It checks for description ambiguity, over-permissive tool grants, and anti-patterns that break headless workflows. Use it when onboarding new agents, integrating agent fleets, or debugging inconsistent multi-agent selection.

This tool audits and validates all Claude Code hooks defined across your settings files. It checks for validity, performance safety, correct event scoping, and proper exit-code handling (blocking vs. non-blocking). Essential for initial setup, debugging hooks that fail to fire, or performing routine system hygiene passes to ensure your agent's tool flow is robust and reliable.

An advanced tool designed to audit, validate, and maintain rule files (e.g., .claude/rules/). It checks structural correctness, validates glob patterns against actual project files, and runs an interactive session to confirm or improve the relevance of each rule. Ideal for initial setup, debugging, or periodic rule hygiene.

This tool audits all existing skills within the current project. It performs a comprehensive quality check, validating frontmatter completeness, tool scoping, and content quality against established standards. It produces a detailed, scored report, including effort-level recommendations for each skill. Ideal for new project onboarding, pre-release quality assurance, or diagnosing skill triggering issues.

This skill evaluates de-identification and clinical NER models against rigorous leakage-first release gates (G1a-G8). Instead of relying on aggregate F1 scores, it focuses on detecting residual Protected Health Information (PHI) leakage and enforcing per-label recall floors, which is crucial for HIPAA compliance. Determines if a model is RELEASABLE or QUARANTINED.

A comprehensive guide for designing and building large-scale, resilient platforms using the evolutionary modular monolith approach. It enforces strong logical boundaries using DDD principles (bounded contexts, flat-by-aggregate) and implements necessary patterns like Anti-Corruption Layers and transactional outboxes. Use this when defining a system's structure, choosing between monolith and microservices, or ensuring robust decoupling from external vendors (e.g., ERP, AI).

This skill provides a structured way to understand complex codebases, functions, or architectural concepts. It allows the user to specify the depth of explanation (Simple, Standard, Deep), ensuring the output covers purpose, step-by-step flow, key decisions, and trade-offs for maximum learning efficiency.

This skill facilitates the extraction of population-scale (cohort-level) clinical text from Electronic Health Records (EHR) or FHIR data warehouses using the FHIR Bulk Data Access ($export) operation. It manages the asynchronous job kickoff, polling, and downloading of NDJSON files. The resulting data stream is designed to be fed in batches into OpenMed for mandatory de-identification and Named Entity Recognition (NER) analysis.

Converts raw OpenMed NER outputs (text spans) into structured, standards-compliant FHIR R4 resources (e.g., Condition, Observation, MedicationStatement). It provides mechanical helpers for building CodeableConcepts using major coding systems like SNOMED, LOINC, and ICD-10, ensuring high interoperability with EHR systems and registries.