Skill UI

Resolve free-text scientific labels to ontology IDs and validate existing CURIEs using EBI OLS4, Bioregistry, and Identifiers.org. Supports annotating biological fields like tissue, disease, and phenotype. Ensures metadata accuracy for submissions to GEO, ENA, and BioSamples by checking obsolescence and prefix correctness.

This skill guides the transformation of a literature review from a mere descriptive catalogue into a highly focused, argumentative structure. It teaches how to synthesize existing research to identify a specific, unresolved theoretical tension or assumption failure within a named scholarly conversation. The goal is to precisely position a new theory or mechanism as the necessary resolution to this gap.

This guide outlines the rigorous standards for quantitative analysis required in academic economic research, particularly dynamic modeling. It provides a framework covering systematic model calibration, moment estimation, numerical solution discipline, and establishing fully reproducible computation workflows, ensuring results are highly auditable and verifiable.

This guide details the critical steps and changes required to transform a reviewed SIGIR submission into the final Version of Record published via ACM Digital Library. It covers technical requirements, metadata standardization (CCS, keywords), conflict resolution, and adhering to publishing timelines and copyright protocols (e-rights, TAPS).

A comprehensive guide for producing the mandatory supplementary video figure for UIST submissions. This material treats the 3-minute video as core evidence, detailing requirements for resolution (1080p/4K), codec (H.264), and anonymity. It guides authors on scripting the video around key claims, ensuring uncut interaction loops, and maintaining academic rigor to convince reviewers of the system's reality.

A robust, on-device solution for processing large-scale clinical text corpora. Supports batch NER, PII extraction, and de-identification over thousands of notes. Features include sharding, checkpointing, resumable functionality, and streaming results to JSONL, ensuring PHI safety and constant memory usage for massive datasets.

This expert, interactive assistant provides comprehensive troubleshooting for complex issues encountered with Claude Code. It systematically scans the user's environment, checks global and project configurations, and cross-references symptoms against a deep knowledge base to pinpoint root causes and provide actionable, step-by-step solutions. Ideal for resolving integration errors, permission issues, or unexpected behavior.

A comprehensive methodology for advanced red teaming and adversary emulation. This skill details how to bypass modern Endpoint Detection and Response (EDR) solutions (e.g., Defender, CrowdStrike, SentinelOne). Techniques include unhooking ntdll, utilizing indirect syscalls (Hell's Gate), patching ETW providers, and AMSI context blinding. Essential for authorized penetration testing and self-developed implant development, aligning with MITRE ATT&CK T1562.

A comprehensive tool designed to help developers compare multiple technical approaches for solving a single problem. It systematically identifies and analyzes the trade-offs, complexity, performance, and learning curves associated with different design patterns (e.g., state management, data fetching, API design). Ideal for architectural reviews and making informed, robust technology stack decisions.

This skill maps clinical concept spans (disorders, anatomy, procedures, etc.) extracted by OpenMed to SNOMED CT concept IDs (SCTID). It performs complex, out-of-process terminology resolution using standard FHIR R4 endpoints ($lookup, $translate, ValueSet/$expand) against a user-supplied, licensed terminology server (e.g., Ontoserver, UTS). Ideal for generating rich, hierarchy-aware clinical codes, performing subsumption queries (ECL), or translating codes between clinical standards (like ICD-10 to SNOMED CT).

This command acts as a crucial product gate, forcing a deep, strategic review before any code is written. Instead of simply fulfilling the request, it guides users through three structured roles: Scope Expansion (finding the 10x vision), Hold Scope (stress-testing assumptions and failure modes), or Scope Reduction (defining the true MVP). It ensures that the developed solution solves the user's real job-to-be-done, not just the explicit feature request.

Performs comprehensive, two-layer validation on generated plans. Layer 1 executes instant structural checks (format, dependencies, compliance) while Layer 2 triggers specialist agents (e.g., security, DB, performance) based on plan content. The system forces resolution of all identified issues before allowing execution, supporting auto-fixes and guided human input.