Produce a complete mobile app architecture design — platform choice, navigation structure, state management, data layer, key screens.
Produce a mobile feature spec — user story, technical approach, component breakdown, platform-specific considerations, edge cases.
Mobile reconnaissance — understand the app's tech stack, architecture, dependencies, and health for takeover.
Use when asked about mobile UI guidelines, touch targets, platform-specific UI rules, or mobile interaction patterns.
Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.
A generic workflow for self-starting TouchDesigner via computer-use (or an already-running instance), loading mcp_webserver_base.tox, and verifying any TD-side Python change…
Use for cognitive task analysis, CTA method selection, knowledge capture, tacit expertise routing, and representation design by matching elicitation methods to knowledge type.
Use when building Towns Protocol bots - covers SDK initialization, slash commands, message handlers, reactions, interactive forms, blockchain operations, and deployment — from…
Use when working on tpcp code that uses n_jobs, joblib parallelism, tpcp.parallel, caching in parallel workers, or when debugging multiprocessing, serialization, or global state…
Hardware meta-skill detailing Google's TPU architecture, specifically the Matrix Multiplication Unit (MXU) and Systolic Arrays.
Instrument agent sessions so operators can replay runs, track model costs, inspect tool calls, and compare behavior across frameworks.
Instrument LLM and agent applications with MLflow tracing, evaluation, prompt tracking, and monitoring so operators can debug behavior before and after deployment.
Instrument Python or TypeScript agent applications with OpenTelemetry traces, metrics, and evaluation views for debugging production LLM workflows.
Applies systematic tracing and isolation techniques to pinpoint exactly where a bug originates in code.
Debug a codegen issue by tracing from wrong Python output back through the graph to the root cause. Identifies whether the bug is in primitive definitions, type resolution, or…
Use when investigating a specific trace_id — following a single request across services to understand where it failed, hung, or behaved unexpectedly.
Safe refactoring workflow using trace-mcp — assess risk, find candidates, check impact, and rename symbols across all files without missing import sites or cross-file references.
Lets an agent record Python allocation traces and inspect the biggest allocators, retained objects, and leak paths before memory growth turns into a production incident.
Trace function call hierarchies, dependencies, side effects, and symbol relationships using LCI. 以LCI追蹤函數調用層次、依賴、副作用、符號關係。 Use when: understanding function before modifying,…
Trace existing execution paths before changing unfamiliar behavior. Use when editing unknown code, debugging, reviewing AI-generated changes, or when a plan depends on current…
Use when working with Tracearr media-server monitoring for Plex, Jellyfin, or Emby, including real-time sessions, stream analytics, account-sharing detection, trust scores,…
Turn any codebase into evidence-grounded Markdown docs plus a machine-readable index.json. Every claim cites its source; never invents deployment steps.
Use when facing technical uncertainty, unproven architecture, or building a large feature where agents or humans risk getting lost in details before confirming the architecture…
Use TokenTracker to auto-collect local token, cost, and rate-limit telemetry across Claude Code, Codex, Cursor, Gemini, OpenCode, OpenClaw, and other AI coding tools.
Retrieve seeded protest, riot, strike, and civil unrest events. Use when the user asks whether social unrest is occurring in a country or over a recent time window.
Track and manage regression test suites across releases. Use when performing specialized testing. Trigger with phrases like "track regressions", "manage regression suite", or…
Evaluate 2–4 architectural options against quality attributes using a weighted decision matrix, quality attribute scenarios, and sensitivity analysis.
Solana swap execution, MEV-protected transaction submission, wallet scanning, and market-cap-based limit/trailing orders via the TradeRouter API.
Traduce textos del español al inglés manteniendo estrictamente el contenido, las ideas y la estructura original sin realizar modificaciones.
Harden a Traefik 2.x/3.x reverse proxy against abusive or unwanted traffic — per-identity rate + concurrency limiting, IP allowlisting, request-body buffering, middleware…
Traefik cloud-native ingress router with automatic TLS, dynamic configuration, middleware chains, and Kubernetes IngressRoute CRDs. Sources: traefik/traefik (MIT).
Build a safe recommendation plan for PlanetScale Postgres Database Traffic Control budgets and rules without applying them.
Trafilatura is a Python package and CLI tool for gathering text from the web. It handles crawling, downloading, and extracting main text content, metadata, and comments from raw…
Train focal and subsidiary attention switching for diagnosis, debugging, design review, and other judgment-heavy work.
Train a character/identity LoRA locally on FLUX.1-dev via the comfyui-mcp train_* tools (GPU Docker + ostris ai-toolkit).
Prepare, launch, and evaluate Kimi LoRA fine-tuning for Smithers TypeScript/JSX workflow authoring. Use when building a Smithers training corpus, validating Fireworks JSONL,…
Split data before looking at it, respect time and groups, and quarantine the test set so scores mean something.
Use when after applying a configuration fix (e.g., adding an instrument type to an allowlist, updating filtering thresholds) to a dataset preprocessing pipeline, you need to…
Train Test Splitter - Auto-activating skill for ML Training. Triggers on: train test splitter, train test splitter Part of the ML Training skill category.
Diagnose NaN losses, out-of-memory errors, and shape mismatches in deep learning or ML pipelines.
Use when after labeling a representative subset of peaks (typically 10–20 pooled samples with corresponding feature tables) and before neural network training, when you need to…
Direct role-archetype mapping, optionality architecture, IDPs and OKRs, skill-gap analysis, and pivot sequencing to the appropriate specialist skill.
Fehleranalyse in Pipeline- und Steuerdatei-Abläufen: prüfen, ob eine Konvention oder ein Verfahren für ein LLM überhaupt sichtbar und auffindbar ist.
Use when reasoning about the I in ACID — the isolation level a database provides between concurrent transactions: the five SQL-standard levels (read uncommitted, read committed,…
Wrap multi-write operations in transactions; use dispatchAfterCommit and idempotency patterns to ensure consistency
Use when implementing attention mechanisms, building custom transformer models, understanding positional encoding, or optimizing transformer inference - covers self-attention,…
Use when when building a neural network to map between mass spectrometry spectra and molecular properties (e.g., fingerprints, SMILES, or fragment ions) where sequential or…
Use when you have acquired or generated multi-modal spectroscopic data (integrated NMR, HSQC, COSY, IR spectra) in the model's expected input format, a pre-trained…
Use when when processing sequential spectroscopic data (1H NMR spectra) where both local chemical shift patterns and global spectral dependencies are needed for compound…
Use when you have paired tandem MS spectra and either (1) molecular fingerprints
Use when you have variable-length MS/MS peak lists (m/z arrays and intensity arrays) that must be fed into a transformer architecture for tasks like compound identification or…
Preflight reference for HuggingFace snapshots — what vLLM, sglang, and transformers.generate see at runtime.
Loading and using pretrained models with Hugging Face Transformers. Use when working with pretrained models from the Hub, running inference with Pipeline API, fine-tuning models…
Run Hugging Face models in JavaScript or TypeScript with Transformers.js in Node.js or the browser.
Use when integrating Hugging Face Transformers.js (Xenova/transformers) for in-browser or in-Node inference, debugging quantized model loading, building bi-encoder / cross-encoder…
Transit Least Squares (TLS) algorithm for detecting exoplanet transits in light curves. Use when searching for transiting exoplanets specifically, as TLS is more sensitiv — from…
Transit Least Squares (TLS) algorithm for detecting exoplanet transits in light curves. Use when searching for transiting exoplanets specifically, as TLS is more sensitiv — from…
Plan and execute language-to-language server or application migrations with behavior parity. Use when porting modules, services, APIs, or runtimes between programming languages;…
Halo 미니앱(`apps/miniapp`)의 다국어 JSON 리소스를 40개 locale 전체에 자연스럽게 번역하는 워크플로우. 새 키/화면 추가 시 모든 locale 동기화. locale 별로 별도 Agent 를 병렬 spawn 해서 WebSearch 로 native 표현을 확인한 뒤 자기 locale json 에…
Behavior-preserving refactoring under characterization tests, plus severity-rated code review. Use when the user says "refactor", "clean up", "tech debt", "code review",…