메뉴
SkillsMP는 Sandeeprdy1729/claude-design-skill에서 22개의 skill을 수집했습니다. skill을 열어 소스와 세부 정보를 확인하세요.
수집된 skill 22개 중 22개를 표시합니다.
Self-critique and Constitutional AI reasoning skill. Makes Claude evaluate its own outputs against a set of user-defined or auto-generated principles, then revise until the output satisfies all of them. Reduces hallucination, over-confidence, and sycophancy…
원문 언어: 영어
Constraint-based problem solving skill. Finds elegant solutions when the design space is heavily restricted — budget caps, time limits, technology mandates, team size, ethical bounds, regulatory requirements. Treats constraints as the design material, not the…
원문 언어: 영어
Cross-domain synthesis skill. Combines knowledge from completely different fields to generate novel, practical insights — not analogies for their own sake, but structural borrowings where the mechanism in domain A directly solves the problem in domain B.…
원문 언어: 영어
Rapid iterative refinement skill. Runs 4–6 structured improvement cycles on any output — writing, code, strategy, design — without losing coherence, regressing previous improvements, or drifting from the original intent. Each cycle targets a specific…
원문 언어: 영어
Large document and codebase synthesis skill. Activates when the user provides 50k–200k+ tokens of context — codebases, PDFs, legal documents, transcripts, research corpora — and needs synthesis, cross-referencing, or insight extraction with perfect recall…
원문 언어: 영어
Multi-step prompt chaining skill where each step's output becomes the next step's input, with explicit rules, constraints, and reference tokens between steps. Activates when the user needs to build a complex pipeline of dependent Claude calls — research →…
원문 언어: 영어
Adversarial red-teaming skill for code, systems, strategies, and plans. Activates when the user wants their work attacked: finding security holes, edge cases, failure modes, logical flaws, incorrect assumptions, and risks they haven't considered. Different…
원문 언어: 영어
Silent failure detection skill. Identifies when Claude is being confidently wrong — surfacing hidden assumptions, exposing overconfident claims, probing calibration, and catching hallucination before it causes damage. Uses specific questioning techniques to…
원문 언어: 영어
Strict structured output enforcement skill. Forces Claude to produce output in exact, machine-readable formats — JSON schemas, YAML, CSV, markdown tables with defined rules, typed lists — with zero deviation, no prose contamination, no invented fields, and no…
원문 언어: 영어
Target voice and style mimicry skill. Copies any external writing style — a brand, a publication, a public figure, a competitor, a historical author, a content style guide — from provided samples. Different from voice-auth, which authenticates the user's own…
원문 언어: 영어
Decision-forcing skill that eliminates hedging and commits to one answer. Activate when the user is asking Claude to choose between options, pick a technology, recommend an approach, or make any comparative judgment — and they want a real answer, not a…
원문 언어: 영어
Merged design engineering skill. Auto-activate on ANY UI/UX/design/frontend/component/animation/interface request. Covers: animation physics, anti-slop rules, motion library, token architecture, polish checklist, typography, color, layout, Framer Motion,…
원문 언어: 영어
Self-healing CLAUDE.md generator and maintenance skill. Activates when the user wants to create, update, audit, or repair CLAUDE.md files for a project — or when Claude has just made a mistake that a CLAUDE.md entry could have prevented. Scans the codebase…
원문 언어: 영어
Persistent context management skill that ends the 'let me explain again' tax. Activates when the user is frustrated by having to re-explain their role, project, stack, or constraints at the start of every session — or when Claude's answers feel generic…
원문 언어: 영어
Meta-skill and skill manager. Activate when the user enables this skill as their primary entry point. Reads each incoming query, selects only the skills and MCP tools relevant to that query, loads them into the active context, executes the work, then unloads…
원문 언어: 영어
Project completion skill that turns half-built things into done things. Activates when the user has an 80%-done project — a SaaS with no landing page, a side project with no README, a blog post that's a brain dump, a feature with no tests — and needs to…
원문 언어: 영어
MCP tool deduplication and unified abstraction layer for multi-server setups. Activate when the user has multiple MCP servers installed and wants to reduce tool noise, detect overlapping tools, understand which tool to use for a given task, or create routing…
원문 언어: 영어
Automated PR review and code quality analysis skill. Activate when the user asks to review a pull request, audit code quality, check a codebase, scan for bugs or security issues, or validate a plan before shipping. Fans out parallel sub-agents — one per…
원문 언어: 영어
Adversarial pre-mortem skill that actively hunts for fatal flaws in plans, proposals, decisions, and strategies. Refuses to be supportive or balanced. Activates when the user wants their plan stress-tested, challenged, or attacked before committing — not…
원문 언어: 영어
Root-cause and question-reframing skill that answers the question behind the question. Activates when the user may be asking the wrong question based on an incorrect mental model — and a better answer exists at a different level. Runs a 3-step check before…
원문 언어: 영어
Skill trigger debugging and description auditor. Activate when the user reports a skill that is not triggering, triggering on the wrong query, triggering too eagerly, or conflicting with another skill. Analyses the skill's description field, tests trigger…
원문 언어: 영어
Writing voice authentication and AI-smell elimination skill. Activates when the user wants AI-generated writing to sound like them specifically — not generic, not polished-AI, not corporate bland. Analyzes a writing sample to fingerprint sentence length,…
원문 언어: 영어