Skill UI

This skill provides a comprehensive guide for manuscript authors targeting the StarSem Conference on Computational Semantics (*SEM). It helps authors determine the conference fit, refine framing, establish the necessary evidence bar, and strategize submissions for topics related to meaning representation, semantic parsing, and inference. It is a specialized venue-selection and re-framing tool for advanced NLP/CS papers.

This skill acts as a comprehensive map for developing clinical and biomedical NLP pipelines using OpenMed. OpenMed is a local-first Python library that performs named-entity recognition (NER), protected health information (PHI) de-identification, and FHIR export, all running fully on-device without cloud calls. Use this guide when starting a new clinical text project, needing to select the appropriate capability (e.g., zero-shot NER, auditing, or FHIR conversion), or requiring an entry point for the core OpenMed features.

A comprehensive, multi-phase methodology for debugging complex software issues. This guide mandates systematic root cause investigation, starting from symptom collection, code path tracing, hypothesis formulation, and deterministic reproduction. It teaches developers to fix the underlying cause, not just the visible symptoms, ensuring highly robust and minimal changes.

This tool analyzes your entire codebase, assessing project identity, team size, test maturity, and existing architectural signals. It then recommends the most appropriate, AI-assisted development methodology stack (e.g., Microservices, Enterprise-Gov) and asks targeted questions to refine the plan, providing a concrete quick start guide.

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 robust pipeline guides development teams through a rigorous, five-phase planning process *before* writing any code. It systematically analyzes PRDs, design specifications (UX, accessibility), and existing architecture to produce a comprehensive implementation plan. Key outputs include Architecture Decision Records (ADRs), detailed task breakdowns, and verification strategies, ensuring all technical and design decisions are documented and scoped.

Automates the deployment process for services by generating and executing necessary shell commands. This skill is ideal for CI/CD pipelines or manual staging deployments, allowing users to target specific services and environments while ensuring safety by requiring script review before execution. Note that deployment will restart running services, causing brief downtime.

This skill processes public Shopify App Store reviews (especially low-star ones) to generate a structured, actionable brief. It categorizes feedback into incident risk, repeated friction, pricing confusion, and feature requests, providing a prioritized P0-P3 assessment for product and support teams. It ensures every piece of feedback maintains its public source link and labeling for human review.

A strategic framework for making decisions when facing extreme uncertainty and unclear outcomes. Effectuation instructs users to start with available means (skills, assets, network) and define an affordable loss limit. Progress is driven by taking small, controlled actions to gain real commitments and information, allowing the ultimate goal to emerge naturally.

This skill assists founders in selecting or validating business, brand, or product names for cross-cultural suitability. It evaluates names regarding pronunciation, meaning, spelling, and cultural connotations in target markets to prevent negative associations. Additionally, it suggests new names rooted in the origin language while ensuring professional appeal. Ideal for startups expanding internationally or rebranding for global audiences.

A read-only repository explorer for cold-start code orientation and cross-file localization. It locates where relevant code lives using broad parallel search and returns only compact path:line citations. Use when the exact file or symbol is not yet known; skip when direct search already works.

Inspect and manage Caveman Cloud experiment lifecycles safely. Use to start, approve, cancel, promote, or roll back experiments with strict evidence gates and read-only enforcement until server-authoritative controls ship.