Claude Skills by jeffreytse

Use when designing any screen layout — to arrange elements so the eye naturally travels from most to least important, reducing the cognitive effort required to parse the page.

Use when reviewing or designing e-commerce flows, subscription sign-ups, cookie consent banners, settings pages, and cancellation flows — to identify and remove deceptive UX patterns that manipulate users against their own interests.

Use when validating a UI design, prototype, or live product with real users to find usability problems before or after launch

Use when starting a design project, commissioning design work, or aligning stakeholders before any visual design begins

Use when writing button labels, error messages, placeholder text, helper text, empty state copy, and tooltip content — to make UI text specific, actionable, and honest.

Use when building or reviewing a web page's HTML structure — to ensure heading hierarchy, landmark regions, and semantic elements convey the correct document outline to screen readers and assistive technology.

Use when designing formative assessment strategies to provide ongoing feedback and guide learning during instruction

Use when creating assessment rubrics for evaluating student work, projects, or performance with clear, objective criteria

Use when designing practice schedules that mix different problem types or topics to improve discrimination and long-term retention

Use when designing study strategies or instructional activities that use testing and recall to strengthen long-term memory

Use when designing a learning or study system that schedules review sessions to maximize long-term retention with minimum time investment

Use when defining or structuring the skills, knowledge, and behaviors required for a role, profession, or learning pathway

Use when designing a course, lesson, module, or training program and need to define what learners will be able to do after instruction

Use when writing learning objectives for a course, lesson, or training program that are measurable and aligned to instructional activities

Use when designing a curriculum, flashcard deck, or study system intended to maximize long-term retention of facts, concepts, or procedures

Use when designing an assessment rubric for a complex task, project, or performance where multiple criteria must be evaluated consistently

Use when adding gamification mechanics to a learning, product, or engagement system to increase motivation, participation, and completion

Use when designing a short, focused learning module (3–10 minutes) targeting a single skill or knowledge application

Use when designing an online course structure including learning objectives, module sequencing, assessment, and delivery format

Use when taking notes during a lecture, seminar, class, or while reading a dense text — to capture information in a structured three-section format that builds in review, retrieval practice, and summarization without extra study sessions.

Use when deciding how to allocate learning time across fields, feeling stuck or stale after deep specialization in one domain, or building a deliberate practice of studying unrelated disciplines to generate transferable insight rather than committing all learning time to a single specialty.

Use when designing a practice regimen to build expert performance in any skill domain — to structure sessions around specific weaknesses, immediate feedback, and focused repetition at the edge of current ability rather than comfortable repetitive performance.

Use when learning a new concept, diagnosing why understanding feels shallow, or preparing to explain complex material — to verify genuine comprehension by explaining the topic in plain language, identify exactly where the explanation breaks down, and fill those gaps until the explanation is clear and jargon-free.

Use when memorizing large ordered sequences — speech outlines, vocabulary lists, historical dates, names, numbers, or any information that must be recalled in sequence without notes.

Use when brainstorming, exploring relationships between concepts, planning a project scope, or creating an overview of a topic — to generate and organize ideas visually as a radial network rather than a linear list.

Use when designing how a new concept or problem-solving skill is introduced — deliberately have learners attempt a problem beyond their current ability BEFORE any direct instruction, then consolidate with explicit teaching that contrasts their failed attempts against the canonical solution, rather than teaching the correct method first and practicing afterward.

Use when reading a textbook chapter, research paper, technical document, or any dense non-fiction text — to maximize comprehension and retention on first pass by converting passive reading into active questioning and retrieval.

Use when facilitating discussion, coaching, or teaching that develops critical thinking through structured questions rather than direct instruction

Use when designing classroom or online sessions that replace passive lecturing with activities requiring learner mental engagement and practice

Use when planning lessons or units that must accommodate learners with diverse readiness levels, learning profiles, or interests within the same classroom or group

Use when designing in-lesson or mid-unit checks that help teachers adjust instruction and help learners self-correct before a summative assessment

Use when redesigning a course so that content delivery happens outside class (video, readings) and class time is used for practice, discussion, and application

Use when facilitating a structured session where learners teach, review, or problem-solve together in pairs or small groups to deepen understanding and develop metacognitive skills

Use when facilitating a learning retrospective after a course, project, or experience to help participants extract and apply lessons

Use when building or tuning full-text search relevance, scoring documents against a keyword query, or building the sparse/keyword leg of a hybrid RAG retriever — including deciding between BM25, raw TF-IDF, and dense/vector search, or diagnosing why keyword search results feel poorly ranked.

Use when deciding whether AI-generated code can be trusted without reading it line by line — instead of reading the target code, build or generate small, deterministic, narrowly-scoped verification tools (coverage checkers, complexity linters, mutation testers) to check it, and trust those tools not by reading them either, but because their small, deterministic scope makes them exhaustively testable by their own dedicated test suite, terminating the trust chain in a genuinely verifiable place.

Use when building or iterating on an LLM feature, prompt, or agentic system — to measure improvement objectively rather than speculating about what the model needs.

Use when rendering LLM-generated text in a browser, executing LLM-generated code, or inserting LLM output into a database or downstream system — any place LLM output leaves the AI layer and enters another execution context.

Use when building applications that make LLM API calls, run inference locally, or build agentic systems — to prevent runaway costs, infinite loops, and denial of service via unrestricted model consumption.

Use when deploying an LLM application that may have been trained on or given access to sensitive data — to assess the risk of the model leaking PII, proprietary information, or confidential context through its outputs.

Use when evaluating, auditing, or improving the quality, factual accuracy, or safety of LLM-generated outputs

Use when deploying ML models that make decisions affecting people (hiring, lending, healthcare, content moderation) or when required by regulation to assess algorithmic bias

Use when curating, collecting, or managing datasets for training or fine-tuning machine learning models — to detect and mitigate poisoned, biased, mislabeled, or adversarially crafted training data.

Use when building or prompting a multi-step agentic system — to structure execution so the agent plans before acting, creates verifiable checkpoints, and avoids irreversible mistakes.

Use when an organization is establishing oversight for how it develops, deploys, or uses AI systems — classifying AI use cases by risk level, assigning accountable owners, and applying oversight proportionate to each use case's actual risk, rather than applying either no governance or the same uniform review process to every AI use case regardless of its actual risk level.

Use when building LLM agents that can take actions in the world — sending emails, modifying files, making API calls, or executing code — to prevent the model from taking actions beyond its intended scope.

Use when building tools, plugins, or function-calling integrations for LLM systems — defining what actions the model can invoke, how those invocations are validated, and how to prevent misuse of plugin capabilities.

Use when building or refactoring machine learning systems that need to move from experimentation to reliable production deployment

Use when training a reinforcement-learning agent where real-environment interaction is expensive, slow, or limited — robotics, game-playing agents, or any control task — by training a model of the environment's dynamics from real interaction data, then planning or training the policy against simulated rollouts from that learned model rather than requiring real-environment interaction for every training step.

Use when building LLM-powered applications that process user input alongside system instructions — chatbots, AI agents, document processors, code assistants, or any system where untrusted text reaches an LLM.