Skill UI

This mode defines an autonomous, senior-level AI agent capable of resolving complex tasks from start to finish with minimal human supervision. It operates by strictly verifying every claim against real evidence (live web search, codebase, available tools). The agent prioritizes action over questioning, only interrupting the user for irreversible actions, genuine dead-ends, or critical ambiguities, ensuring maximum efficiency in hands-on engineering and operational workflows.

This skill processes clinical text to interpret laboratory results. It parses reference ranges (e.g., "135-145", "<5") and derives structured abnormal flags (low, normal, high, critical) for a given numeric value. It is unit-agnostic and respects explicit flags provided by the originating lab, making it crucial for structured clinical data extraction and interpretation workflows.

This skill provides a comprehensive, 4-phase workflow for managing and reviewing a pull request backlog. It automatically audits open PRs, classifying them by size, staleness, overlap, and CI status. Users can then initiate deep code reviews, generate comment drafts, and set up local worktrees for systematic maintenance and quality assurance.

A comprehensive, multi-phase workflow guide for authorized reverse engineering of complex network protocols. Ideal for analyzing proprietary formats such as custom TCP/UDP binary protocols, Protobuf, gRPC, WebSocket frames, and reconstructing state machines from raw PCAP data. This guide covers data collection, field layout restoration, serialization recovery, and secure replaying.

A robust template designed for creating complex, multi-stage commands or 'recipes.' It enforces a highly structured process by mandating 'Context Validation Checkpoints' before any step executes. This pattern ensures that preconditions are explicitly verified, significantly increasing the reliability and predictability of complex automated workflows.

This skill facilitates reversible de-identification (pseudonymization) of clinical text, allowing sensitive Protected Health Information (PHI) to be masked for processing while retaining the ability to restore the original data later under strict authorization. It captures a mapping key and provides a function to re-identify the text, crucial for workflows like GDPR compliance, patient recontact, or auditing. Crucially, the re-identification mapping must be stored separately and securely.

This tool analyzes a given codebase to automatically identify high-value, actionable automation routines. It surfaces use cases across three primary trigger types: scheduled cron jobs, external API calls, and GitHub repository events. By mapping manual pain points to autonomous workflows, it helps teams implement robust CI/CD and DevOps practices.

This skill guides users on deploying OpenMed models completely on-device for maximum privacy and edge computing. It details running workflows using three backends: MLX (Apple Silicon), CoreML (iOS/macOS), and ONNX/WebGPU (cross-platform/browser). The process involves converting, optionally quantizing, and running models locally, ensuring PHI never leaves the host device.

Utilize zero-shot NER to extract custom entity types (e.g., Drug, Symptom, Device) from complex clinical or biomedical text without requiring any labeled training data or fine-tuning. OpenMed wraps GLiNER/GLiNER2, allowing users to define labels on the fly. The workflow involves building a model index and then running inference via CLI or Python API, making it ideal for rapid prototyping or unstable schemas.

This command inspects the native code execution sandbox environment. It reports on platform compatibility (e.g., bubblewrap, Seatbelt), active security configurations (auto-allow status, filesystem policies, network allowlists), and details recent security violations. Use it to audit the execution environment before running autonomous workflows.

A comprehensive guide for embedding OpenMed-powered tools into Electronic Health Records (EHRs) like Epic or Cerner. It covers the entire SMART on FHIR workflow, including OAuth2 PKCE authorization, granular scoping (e.g., patient/DocumentReference.rs), secure token handling, and processing clinical notes locally on-device for de-identification and Named Entity Recognition (NER).

This orchestrator manages the entire end-to-end pipeline for preparing technical talks, transforming raw materials—whether code commits (REX mode) or conceptual articles—into polished revision sheets. It sequentially runs through multiple stages (Research, Concepts, Scripting, Revision) with human-in-the-loop checkpoints, ensuring a structured and comprehensive content creation workflow.