Argue against the strongest version of opposing positions, not the weakest. Use for design reviews, evaluating alternatives, conflict resolution, and decision validation.
Identify and manage the bottleneck; improvements elsewhere don't matter until the constraint is addressed.
Use when facing a decision, unclear or recurring problem, system behavior question, communication challenge, or when generating novel solutions — loads structured thinking…
Apply TRIZ (Theory of Inventive Problem Solving) methodology to resolve technical contradictions and find innovative solutions.
Improve by removal rather than addition. Focus on what to stop doing, eliminate the negative, and subtract complexity.
Use when a project CLAUDE.md is bloated, predates the canonical skeleton, or duplicates the shared home layers (the ai/core behaviour-protocol and the personal…
ThinQ integration. Manage data, records, and automate workflows. Use when the user wants to interact with ThinQ data.
Drafts a Third-Party Complaint (impleader) under FRCP 14 or state equivalents. Use when a defendant needs to implead a party for indemnification, contribution, subrogation, or…
Use when reviewing a website for privacy compliance, third-party resource loading, or cookie consent implementation.
Integrate external APIs and services with error handling, retry logic, and data transformation. Use when connecting to payment processors, messaging services, analytics platforms,…
Prüft Streitverkündung, Nebenintervention und Drittbeteiligung bei englischsprachigen Commercial-Court-Verfahren.
Review third-party skill repos for format fit, provenance, and first-party adoption paths without treating them as direct harness dependencies.
Drafts irrevocable third-party Special Needs Trusts funded by parents, grandparents, or other non-beneficiary parties to supplement government benefits without triggering Medicaid…
Überprüft Third-Party-Agenten und Intermediäre auf FCPA/UK-Bribery-Risiken – Due Diligence, Verträge, Red Flags und Vertragsbeendigung im Internal Investigations Praxis.
Persist durable knowledge when it would change how a future session behaves. Capture atomic, deduplicated vault notes or route actionable code rules to the appropriate rules file.
Provides guidance on creating custom error types with thiserror, including proper derive macros, error messages, and source error chaining.
Write Go code in the style of Ken Thompson, co-creator of Go, Unix, and C. Emphasizes Unix philosophy, minimalism, and programs that do one thing well.
Entry point and router for THOR-related work: running scans, analyzing THOR logs, troubleshooting THOR behavior, maintaining THOR installs, THOR Lens workflows, writing THOR…
GSD 계획 생성 + Architect 검토 + 병렬 실행을 자동으로 수행. "꼼꼼하게", "thorough", "careful phase" 요청 시 사용.
Use for any request to review, rate, score, assess, evaluate, grade, benchmark, or otherwise judge quality; applies a custom weighted scoring model with multiple checks per…
Score every decision point with a Thoroughness Rating (1-10). AI makes the marginal cost of doing things properly near-zero — pick the higher-rated option every time.
The Thorsten Prompt — a prompt coach built from Thorsten Ball's (Amp) playbook. Grades every non-trivial request against the Slack test and three checks (standard, information,…
思想演化教练技能 — 帮助用户追溯思想的演化脉络，将分散的碎片意识合并为系统化的高阶认知结构。Thought evolution coaching skill — helps users trace thought lineage and merge scattered fragments into higher-order cognitive…
Plant Alerts, Webinare, Beiträge und Mandantengespräche mit Nutzen. Für Managing Partner, Management Committee oder COO/CFO einer deutschen mittelständischen Kanzlei mit konkreten…
Visualizes Product Manager thoughts (Why, What, How, User Journey) into an editable Excalidraw diagram.
Your thoughtful companion for WhatsApp - remembers what matters, helps you stay present in your relationships.
Use for planning and implementing features with strong end-user orientation by identifying expected core capabilities, practical nice-to-haves, and scope-safe enhancements that…
Thoughtful GPT integration. Manage data, records, and automate workflows. Use when the user wants to interact with Thoughtful GPT data.
Thoughtly integration. Manage data, records, and automate workflows. Use when the user wants to interact with Thoughtly data.
ThoughtSpot integration. Manage data, records, and automate workflows. Use when the user wants to interact with ThoughtSpot data.
Convert a ThoughtSpot model/worksheet and its Liveboards into a Sigma data model and matching dashboards.
Use before posting anywhere outside this task's own thread, or when deciding how much to post inside it.
Keep one Claude Code thread alive across context limits and across the user's own Claude accounts. Use when the user asks to rotate/refresh the session or free up context, for the…
Create multi-tweet X threads that hook hard, deliver escalating value across 6-10 tweets, and close with a clear CTA — optimized for X's format, algorithm, and reading behavior.
Manage X thread structures — hook tweet first, proper sequencing, auto-plug link timing (after X likes or Y minutes), visual placement, and cross-tweet flow.
Structured X thread templates — 12 proven thread formats (listicle, story, how-to, contrarian, etc.) with fill-in frameworks.
Land a sprawling chat thread on an actual decision — the summarize-and-fork move (positions restated, the question isolated), the decider-and-deadline injection, and the recorded…
Turn a REAL Slack thread (or channel) into a logged decision — read it, extract what was decided, who owns what, and record it in Notion — not a template for writing decisions.
Convert a pasted product, marketing, or engineering discussion into a scoped implementation spec with acceptance checks and explicit out-of-scope boundaries.
Use when when deploying Mass2SMILES inference on CPU-only hardware (e.g., when GPU/CUDA support is unavailable), you need to verify that the InferenceModel cpu_threads parameter…
PyQt/PySide6 threading and concurrency - QThread, QThreadPool, QTimer, thread safety, concurrent patterns
Set up or plan a ThreadOps monitoring workflow for recurring checks, deploy follow-up, CI watch, long-running jobs, reminders, or thread automations.
Use when the user wants to start, reply to, or read a thread between Claude Code sessions, or asks about cross-session async coordination.
Three.js and WebGL adapter patterns for HyperFrames. Use when creating deterministic Three.js scenes, WebGL canvas layers, AnimationMixer timelines, camera motion, shader — from…
Use when the user wants to work with the three-audiences documentation paradigm. Triggers on drift detection ("detect drift", "check drift", "align docs", "are docs up to date"),…
Three.js performance optimization and best practices guidelines. Use when writing, reviewing, or optimizing Three.js code.
Auto-routes work to GPT-5.5 (Codex) or Gemini 2.5 Pro when the task fits their unique capabilities better than Claude alone.
three_dart loaders — GLTFLoader, loading GLTF/GLB models, textures, async patterns, Flutter asset loading.
Plan the React Three Fiber component shape (props, refs, Suspense, useGLTF cache, animation hooks) before delivering a GLB to a web team.
three-globe' v2.x changed atmosphere/globe configuration from method chaining to property setters — `.atmosphereDayIncrement is not a function` at runtime
Map a scoped topic across three coexisting horizons — current system under strain (H1), preferred future (H3), and transition innovations sorted as H2+ or H2- — to understand…
Model extension lifecycle as intent, materialization, and active runtime layers, then sync each layer with dedicated transitions.
Parallel Architect→Builder→Reviewer pattern for complex multi-phase tasks. Dispatches 3 agents: Architect (Opus) reads codebase and writes a plan with phases; Builder (Sonnet)…
3-MATCH gadgets and non-backtracking geodesics for 3-SAT via colored subgraph isomorphism.
三个能干活的 AI 分身协作系统。用 Claude Code 启动三个不同人设的 coding agent，共享同一个工作目录，多轮协作直到达成共识。适用于代码审查、方案评审、重构任务等需要多视角协作的场景。
Design a Montessori three-period lesson for introducing concepts through concrete materials and naming.
Read papers by Keshav's three-pass method and reach a conclusion you can defend. Run when given papers — PDF, arXiv link, or full text — and asked which of them are worth reading,…
Activate when: user says 'I know what to do but I'm not doing it', 'we understand the problem but can't fix it', 'effort isn't translating to results', 'why isn't our team…
The assertion → judge → human eval ladder per Husain & Shankar (Jan 15 2026). No higher tier without the lower tier populated.
三级记忆管理系统 (Three-Tier Memory)。短期记忆滑动窗口 + 中期记忆自动摘要(LLM) + 长期记忆向量检索(RAG)。当需要管理对话历史、自动摘要、语义检索、构建个人知识库时使用此 Skill。