Provides comprehensive code review capability for Next.js applications, validates Server Components, Client Components, Server Actions, caching strategies, metadata, API routes, middleware, and performance patterns. Use when reviewing Next.js App Router code changes, before merging pull requests, after implementing new features, or for architecture validation. Triggers on "review Next.js code", "Next.js code review", "check my Next.js app".

Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution. Use when needing to run interpretability experiments on massive models (70B+) without local GPU resources, or when working with any PyTorch architecture.

Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution. Use when needing to run interpretability experiments on massive models (70B+) without local GPU resources, or when working with any PyTorch architecture.

Implements OAuth 2.0 and OpenID Connect authentication flows with proper security, token management, and common provider integrations.

OBLITERATUS: abliterate LLM refusals (diff-in-means).

Adds tracing, telemetry, and observability to an assistant-ui backend. Use when wiring an AI SDK route handler (streamText/generateText, toUIMessageStreamResponse) to a tracing backend: Langfuse via OpenTelemetry (LangfuseSpanProcessor and NodeSDK in instrumentation.ts, experimental_telemetry isEnabled, propagateAttributes with traceName/userId/sessionId, langfuseSpanProcessor.forceFlush on serverless), LangSmith via wrapAISDK(ai) from langsmith/experimental/vercel (createLangSmithProviderOptions, awaitPendingTraceBatches), or Helicone via createOpenAI baseURL https://oai.helicone.ai/v1 with the Helicone-Auth header. Also covers rendering collected spans with @assistant-ui/react-o11y headless primitives (SpanResource, SpanPrimitive Root/Indent/CollapseToggle/StatusIndicator/TypeBadge/Name/Children, SpanByIndexProvider, SpanData/SpanState) mounted via useAui/AuiProvider from @assistant-ui/store. Use for missing or empty traces, edge vs nodejs runtime telemetry, serverless flush issues, or trace waterfalls.

Microscopy data management platform. Access images via Python, retrieve datasets, analyze pixels, manage ROIs/annotations, batch processing, for high-content screening and microscopy workflows.

Integrate Open-Meteo Weather Forecast, Air Quality, and Geocoding APIs: query design, variable selection, timezone/timeformat/units, multi-location batching, and robust error handling. Use when fetching weather forecasts, air quality/pollen data, or geocoding place names to coordinates via Open-Meteo. Keywords: Open-Meteo, /v1/forecast, /v1/air-quality, geocoding-api, hourly, daily, current, timezone=auto, timeformat=unixtime, models, WMO weather_code, CAMS, GeoNames, httpx, FastAPI, pytest.

OpenClaw session relay: prompts/posts via local/remote acpx over SSH.

Orchestrate Chinese LLMs (DeepSeek, Qwen, Yi, Moonshot) through OpenRouter API with LangChain. Use when: openrouter, chinese llm, deepseek, qwen, moonshot, yi model, model routing, auto router, llm orchestration.

Set up and configure wallet signing providers for OpenSea transactions. Supports Privy, Turnkey, Fireblocks, Bankr, and local private keys. Required for executing trades (opensea-marketplace) and token swaps (opensea-swaps).

Use these skills to manage and monitor Oracle databases by executing SQL statements, exploring schema metadata, analyzing query performance, monitoring active sessions and resource consumption, and managing storage and object health.

Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers, vLLM), and maximize inference speed with Outlines - dottxt.ai's structured generation library

Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers, vLLM), and maximize inference speed with Outlines - dottxt.ai's structured generation library

Binance P2P trading assistant for natural-language queries about P2P/C2C market ads, the user's own P2P order history, order detail & appeal tracking, and advertisement publish & management. Use when the user asks about P2P prices, searching/choosing ads, comparing payment methods, reviewing P2P order history, checking order detail/appeal status, querying complaints, publishing/updating/managing P2P advertisements, or viewing merchant profiles. Do NOT use for spot/futures prices, exchange trading, deposits/withdrawals, on-chain transfers, or anything unrelated to P2P/C2C.

Power systems analysis and optimization library (pandapower v3.4.0). Use when working with electric power networks: building network models (buses, lines, transformers, loads, generators), running AC/DC power flow, optimal power flow (OPF), short circuit calculations (IEC 60909), state estimation, time series simulations, network topology analysis, or visualizing power grids in Python.

Create new agents in Paperclip with governance-aware hiring. Use when you need to inspect adapter configuration options, compare existing agent configs, draft a new agent prompt/config, and submit a hire request.

Pay HTTP 402 payment challenges using tokens via the Tempo CLI and Uniswap Trading API. Use when the user encounters a 402 Payment Required response, needs to fulfill a machine payment, mentions "MPP", "Tempo payment", "pay for API access", "HTTP 402", "x402", "machine payment protocol", "pay-with-any-token", "use tempo", "tempo request", or "tempo wallet".

This skill should be used when the user asks to "split PDF by chapters", "divide book into chapters", "extract chapters from PDF", or "break PDF into sections". Splits PDF documents based on table of contents or text patterns using pypdf.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

This skill should be used when the user says "create a PEP", "update PEP", "PEP status", "document this", "spike", "investigate", "explore", "look into", "scout", "research", "scope out", "figure out", or references PEP-*.md files. Manages PEP (Project Enhancement Proposal) documents covering both investigation ("I don't know the shape yet") and execution ("here's the plan"). Wave shapes, mandatory adversarial review before execution, and evidence-required wave completion. The markdown file is the source of truth; an audit hook records what actually ran.

Raspberry Pi system administration. Monitor resources, manage services, perform updates and maintenance.

Pino high-performance JSON logger for Node.js with worker thread transports, child loggers, redaction, and framework integrations. Use when setting up structured logging, configuring log transports, adding request correlation IDs, redacting sensitive data, or integrating with Fastify, Hono, or Express. Use for pino, logging, structured-logs, request-id, correlation, redaction, transports, pino-http, pino-pretty.

Expert platform and API product management guidance for developer-focused products. Use when planning API product strategy, designing APIs, improving developer experience (DX), creating developer documentation, building SDKs, planning API versioning and deprecation, building developer communities, creating integration marketplaces, or measuring platform health. Covers REST, GraphQL, webhooks, and platform ecosystems.

Automated regression testing strategy and best practices using Playwright with TypeScript. Use when asked to plan, organize, select, execute, or optimize regression test suites for web applications. Covers change-based and risk-based test selection, test tagging and prioritization, parallel execution, sharding, CI/CD pipeline integration with GitHub Actions, flaky test management, suite health monitoring, and regression types (corrective, progressive, selective, complete). Keywords: regression testing, test selection, smoke tests, test suite optimization, CI pipeline, flaky tests, test sharding, impact analysis, git diff.

Interactive scientific and statistical data visualization library for Python. Use when creating charts, plots, or visualizations including scatter plots, line charts, bar charts, heatmaps, 3D plots, geographic maps, statistical distributions, financial charts, and dashboards. Supports both quick visualizations (Plotly Express) and fine-grained customization (graph objects). Outputs interactive HTML or static images (PNG, PDF, SVG).

When the user wants to define their ideal customer profile, position an AI product, build messaging architecture, or validate product-market fit. Also use when the user mentions 'ICP,' 'ideal customer profile,' 'positioning,' 'PMF,' 'product-market fit,' 'messaging,' 'buyer persona,' 'enrichment signals,' 'market positioning,' or 'competitive positioning.' This skill covers market positioning, ICP definition, messaging architecture, and PMF validation for AI-native products. Do NOT use for technical implementation, code review, or software architecture.

Research and qualify onboarding team referral leads for PostHog. Use this skill when a TAE receives a lead from the onboarding team and needs a full research brief before deciding how to engage. Triggers on 'research this onboarding lead', 'onboarding team referred [company]', 'look into [company] from onboarding', 'qualify this onboarding referral', 'what do we know about [company] from onboarding', or any request to research a company that came through the onboarding pipeline. Also trigger when a TAE pastes a company name and mentions it's from the onboarding team, or says something like 'onboarding sent me [company]', 'got a handoff for [company]', or '[name] from onboarding sent me [company]'. This skill does deep research and qualification, then drafts outreach when the recommendation is to engage.

Qualify and draft outreach for PostHog product-led leads who are hitting a billing transition — either startup program customers rolling off free credits, or users whose first invoice will be >= $2K. Use this skill when a TAE needs to work a startup rolloff lead, a high first-invoice alert, or any lead where the matching criteria mentions 'startup rolloff', 'credit spend', 'first invoice', 'credits expiring', or 'high usage transition'. Also trigger when a TAE pastes Salesforce lead details with matching criteria like 'Startup rolloff + high credit spend' or 'First invoice >= $2K'.

Use when setting up Postmark webhooks for tracking email delivery, bounces, opens, clicks, spam complaints, or subscription changes — includes webhook configuration, payload handling, and security.

Expert PR and media relations guidance for earned media, press coverage, and reputation building. Use when writing press releases, crafting media pitches, developing journalist relationships, planning embargo strategies, managing crisis communication, placing thought leadership, building analyst relations, submitting award applications, or measuring PR coverage. Use for media outreach, press kits, speaking opportunities, and earned media strategy.

Pre-commit hooks framework for multi-language code quality automation. USE WHEN setting up pre-commit OR configuring git hooks OR adding linting OR code formatting OR security scanning OR Terraform validation OR Kubernetes manifests OR Helm charts OR Python linting OR JavaScript formatting. Manages .pre-commit-config.yaml, hook installation, and CI integration.

Use when building creative browser demos with @chenglou/pretext — DOM-free text layout for ASCII art, typographic flow around obstacles, text-as-geometry games, kinetic typography, and text-powered generative art. Produces single-file HTML demos by default.

Expert pricing strategy and monetization guidance for SaaS and digital products. Use when designing pricing tiers, creating pricing pages, setting value metrics, developing bundle strategies, planning price increases, writing pricing copy, analyzing competitive pricing, or optimizing annual vs monthly positioning. Use for packaging decisions, discount policies, enterprise pricing, and behavioral pricing tactics.

Expert product launch strategist for SaaS and technology companies. Use when planning product launches, coordinating cross-functional launch teams, managing beta programs, creating launch communication plans, planning launch day execution, setting up post-launch monitoring, running launch retrospectives, or defining launch metrics. Covers launch tiering, internal enablement, rollback planning, and contingency strategies.

Game code architecture, design patterns, scalable systems, and maintainable code structure for complex games.

Use this skill to migrate existing projects to the AgenticDev structure. It uses an AI-powered analysis to intelligently discover, categorize, and migrate documentation, generate rich frontmatter, and preserve git history.

Break down large projects into dependency-aware tasks with progress tracking and visualization. This skill should be used when users want to plan complex implementations, create task breakdowns with dependencies, or visualize project structure.

Creates, optimizes, and iteratively refines agent prompts, system prompts, developer prompts, and reusable prompt templates. Use when asked to improve a prompt, optimize a system prompt, rewrite an agent prompt, tune prompt wording, make a prompt more reliable, port prompts between OpenAI, Claude, or Gemini, or build prompt evals.

Testing patterns for Prowler API: JSON:API, Celery tasks, RLS isolation, RBAC. Trigger: When writing tests for api/ (JSON:API requests/assertions, cross-tenant isolation, RBAC, Celery tasks, viewsets/serializers).

Analyzes pseudocode, algorithms, or code snippets to extract and document functional requirements and business specifications. Produces functional specifications, business logic documentation, data requirements, and workflow specifications focused exclusively on WHAT the system does, not HOW it's implemented. Use when analyzing pseudocode for business logic, extracting functional requirements from algorithms, documenting business behavior from code, reverse engineering functional specifications, or when users mention "pseudocode to spec", "extract requirements", "business logic documentation", "functional specification from code", or "requirements from pseudocode".

Python interface to OpenMS for mass spectrometry data analysis. Use for LC-MS/MS proteomics and metabolomics workflows including file handling (mzML, mzXML, mzTab, FASTA, pepXML, protXML, mzIdentML), signal processing, feature detection, peptide identification, and quantitative analysis. Apply when working with mass spectrometry data, analyzing proteomics experiments, or processing metabolomics datasets.

Genomic file toolkit. Read/write SAM/BAM/CRAM alignments, VCF/BCF variants, FASTA/FASTQ sequences, extract regions, calculate coverage, for NGS data processing pipelines.

Python patterns for system reliability — background jobs and task queues (Celery, async), resilience and recovery (retries, backoff, timeouts, circuit breakers via tenacity), and observability (structured logging via structlog, metrics, distributed tracing, golden signals). USE WHEN building async workers, queueing tasks, handling transient network/IO failures, instrumenting Python services for production, designing retry policies, configuring logging or tracing, or any combination of these system-reliability concerns. NOT FOR language idioms or type hygiene (use `writing-python`) or project setup and dependency management (use `uv`).

High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use when you want clean training loops with built-in best practices.

High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use when you want clean training loops with built-in best practices.

High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use when you want clean training loops with built-in best practices.

Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing, activation patching, interchange intervention training, or testing causal hypotheses about model behavior.

Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing, activation patching, interchange intervention training, or testing causal hypotheses about model behavior.

Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers.

Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers.

Query, filter, and transform Markdown structurally with mq — a jq-like CLI for Markdown. Use to extract headings/sections/code-blocks/links from .md files, build a table of contents, pull code blocks of a given language, slice or reshape LLM prompt/output Markdown, or batch-transform docs. Triggers on "extract sections from this markdown", "get all the code blocks", "jq for markdown", "mq", or any structural query over Markdown that grep/Read can't do cleanly.

R statistical analysis for publication-ready sociology research. Guides you through phased workflows for DiD, IV, matching, panel methods, and more. Use when doing quantitative analysis in R for academic papers.

Expert guidance for building with Radix UI -- Primitives (unstyled accessible components) and Themes (pre-styled design system). Covers composition patterns, asChild prop, accessibility, animation, theming, color system, and keyboard navigation. Trigger when working with Radix components, @radix-ui packages, or building accessible UI primitives. Also trigger on mentions of "radix", "radix-ui", "radix primitives", "radix themes", "@radix-ui/react-*", or the unified "radix-ui" package. TRIGGER WHEN: working with Radix components, @radix-ui packages, or building accessible UI primitives DO NOT TRIGGER WHEN: the task is outside the specific scope of this component.

Comprehensive RAG development knowledge base covering chunking, embeddings, vector databases, retrieval strategies, advanced patterns (Graph RAG, CRAG, Self-RAG, Agentic RAG), evaluation, and production deployment. TRIGGER WHEN: building, implementing, writing, coding, creating, optimizing, or auditing RAG systems. DO NOT TRIGGER WHEN: the task is outside the specific scope of this component.

# Rails Framework Skill - Quick Reference

Use when investigating and documenting a production incident, outage, data corruption event, or post-mortem — guides evidence collection during the investigation AND produces a rich, reproducible Root Cause Analysis report. Trigger on phrases like "write an RCA", "post-mortem for X", "document this incident", "what went wrong with...", "the pipeline broke yesterday, help me investigate", or any time the user is debugging a recently-resolved incident and wants a writeup. Also use proactively when the user finishes resolving an incident in-session and the resolution context is fresh — offer to capture it as an RCA before details fade.

React/TypeScript frontend implementation patterns. Use during the implementation phase when creating or modifying React components, custom hooks, pages, data fetching logic with TanStack Query, forms, or routing. Covers component structure, hooks rules, custom hook design (useAuth, useDebounce, usePagination), TypeScript strict-mode conventions, form handling, accessibility requirements, and project structure. Does NOT cover testing (use react-testing-patterns), E2E testing (use e2e-testing), or deployment.

Analyze real estate and infrastructure investments including REITs, direct property valuation, and infrastructure assets. Use when the user asks about real estate investing, REITs, cap rates, NOI, FFO, AFFO, property valuation, or infrastructure investments. Also trigger when users mention 'rental property analysis', 'cash-on-cash return', 'gross rent multiplier', 'REIT dividends', 'real estate sectors', 'cell towers', 'toll roads', 'LTV ratio', 'DSCR', or ask whether to invest in real estate directly or through REITs.

Protocol for cognitive pattern handovers during complex reasoning sessions. Defines .reasoning/ directory structure, handover schemas, and IR-v2 orchestration integration. Use when reasoning sessions require mid-stream pattern transitions, parallel branch merging, state checkpointing, or multi-pattern orchestration. Essential for complex problems where ToT, BoT, HE, or other patterns must hand off work to each other.

Use when needing to retrieve information from past sessions, conversations, or stored knowledge. Triggers on questions that could benefit from historical context including user preferences, past discussions, workflows, coding patterns, or any query where previous session information would be helpful. Activates broadly for retrieval queries.

Designs self-referential and recursive systems that examine, modify, or generate themselves, including metacognitive architectures and strange loops.

Coordinates behavior-preserving code refactors. Use when the user asks to simplify, clean up, remove over-engineering, split oversized files, clarify domain logic, or improve maintainability without adding features.

Patterns for safe behavior-preserving refactoring — extract, rename, move, simplify — with test-baseline and rollback guidance. Trigger on refactor or restructure requests.

Reinforcement Learning best practices for Python using modern libraries (Stable-Baselines3, RLlib, Gymnasium). Use when: - Implementing RL algorithms (PPO, SAC, DQN, TD3, A2C) - Creating custom Gymnasium environments - Training, debugging, or evaluating RL agents - Setting up hyperparameter tuning for RL - Deploying RL models to production

Manage music releases using RELEASE.md documents stored in releases/{release-slug}/ within an artist workspace. Triggers when the user mentions an artist's album, EP, single, or project — or asks about release planning, DSP pitches, metadata, marketing, press materials, physical production, or tour coordination. First infer which artist and release the user means, then find or create the RELEASE.md. Use this skill to create, update, or pull data from release documents, and to generate deliverables like DSP pitches, press one-sheets, and production specs.

Conducts comprehensive requirements review including completeness validation, clarity assessment, consistency checking, testability evaluation, and standards compliance. Produces detailed review reports with findings, gaps, conflicts, and improvement recommendations. Use when reviewing requirements documents (BRD, SRS, user stories), validating acceptance criteria, assessing requirements quality, identifying gaps and conflicts, or ensuring standards compliance (IEEE 830, INVEST criteria). Trigger when users mention "review requirements", "validate requirements", "check requirements quality", "find requirement issues", or "assess BRD/SRS quality".

Master requirements gathering, user story writing, acceptance criteria definition, and scope management. Transform insights into clear, actionable specifications.

Creates formal academic research papers following IEEE/ACM formatting standards with proper structure, citations, and scholarly writing style. Use when the user asks to write a research paper, academic paper, or conference paper on any topic.

This skill should be used when the user wants to analyze evaluation patterns across completed plans and evolve checklists. Triggered by asking to "run a retrospective", "analyze evaluation patterns", "evolve checklists", or "/superpowers:retrospective".

Interactive post-session retrospective that captures learnings, updates skills, and saves memories. Use when the user says "/retrospective", "let's do a retro", "what did we learn", "session review", "retro", or "wrap up". Also use at the end of long productive sessions when significant patterns or corrections emerged. Supports multi-session mode — by default processes all of today's sessions across projects.

Pre-landing PR review. (gstack)

Perform comprehensive multi-agent code review for PRs, commits, or entire codebases. This skill should be used when a user wants a thorough code review covering correctness, performance, code style, test coverage, and error handling. Analysis only - identifies issues without modifying code.

/rev - Senior Full-Stack Code Reviewer with 12+ years experience in Java/Kotlin and TypeScript/React. Use when reviewing code quality, checking security vulnerabilities, validating style compliance, running static analysis tools, ensuring test coverage, or verifying implementation against acceptance criteria. Also responds to 'Rev' or /rev or /reviewer command.

AMD ROCm GPU computing stack for HIP kernel development and GPU-accelerated library usage. Use when: writing HIP kernels (.hip files), using rocBLAS/rocFFT/rocRAND/rocSOLVER/rocSPARSE/hipBLAS/hipBLASLt/hipTensor/hipSPARSELt/rocALUTION compute libraries, profiling with rocProfiler or rocprof, porting CUDA code to HIP, building CMake/Makefile projects targeting AMD GPUs, using HIP Graphs for low-overhead kernel replay, or debugging GPU code with rocGDB.

Orchestrate multiple code-writing agents in parallel when the work contains 3 or more genuinely independent build tasks. Use for execution lanes with disjoint write scope, clear task boundaries, and no dependency edges that would force sequencing.

Orchestrate multiple debugging agents in parallel when there are 3 or more genuinely independent failures, bugs, or broken subsystems. Use for isolated debug lanes with disjoint root causes, disjoint write scope, and no sequencing requirements.

Rust testing conventions and frameworks: built-in #[test], integration, and doctest layout plus the cargo-nextest runner and the proptest, insta, criterion, mockall, and rstest ecosystem. Invoke whenever task involves any interaction with Rust tests — writing, running, configuring, or debugging tests in .rs files and Cargo projects.

RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production at Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters.

RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production at Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters.

Safely analyze and reorganize project structure with multi-stage validation, dry-run previews, and explicit user confirmation. Use when projects need cleanup, standardization, or better organization.

Saleor e-commerce entity types, relationships, and identifier rules. Use before answering any question about Saleor entities, how they connect, identifier rules (slug vs name), channels, or what Configurator manages.

CRM integration, lead tracking, outreach automation, and pipeline management. Transform your AI agent into a sales assistant that never lets leads slip through the cracks.

Capture desktop screenshots with OS-aware save-location rules, app/window/region support, and bundled helpers.

Create a complete Sentry SDK skill bundle for any platform. Use when asked to "create an SDK skill", "add a new platform skill", "write a Sentry skill for X", or build a new sentry-<platform>-sdk skill bundle with wizard flow and feature reference files.

When the user wants to audit, review, or diagnose SEO issues on their site. Also use when the user mentions "SEO audit," "technical SEO," "why am I not ranking," "SEO issues," "on-page SEO," "meta tags review," "SEO health check," "my traffic dropped," "lost rankings," "not showing up in Google," "site isn't ranking," "Google update hit me," "page speed," "core web vitals," "crawl errors," or "indexing issues." Use this even if the user just says something vague like "my SEO is bad" or "help with SEO" — start with an audit. For building pages at scale to target keywords, see programmatic-seo. For adding structured data, see schema. For AI search optimization, see ai-seo.

Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs, constrained decoding, agentic workflows with tool calls, or when you need 5× faster inference than vLLM with prefix sharing. Powers 300,000+ GPUs at xAI, AMD, NVIDIA, and LinkedIn.

Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs, constrained decoding, agentic workflows with tool calls, or when you need 5× faster inference than vLLM with prefix sharing. Powers 300,000+ GPUs at xAI, AMD, NVIDIA, and LinkedIn.

Generate correct shadcn/Tailwind layouts by applying CSS mental models. This skill should be used when the user asks to 'create a shadcn layout', 'fix layout issues', 'debug CSS height problems', 'make scrolling work', or has issues with Tailwind flex/grid. Keywords: shadcn, Tailwind, layout, CSS, flex, grid, height, scroll, overflow.

Shell script conventions, defensive patterns, and correctness rules: strict mode, quoting, portability, error handling, and common pitfalls. Invoke whenever task involves any interaction with shell scripts — writing, reviewing, debugging, or understanding .sh, .bash, .zsh files.

Advanced theming for Shiny apps using bslib and Bootstrap 5. Use when customizing app appearance with bs_theme(), Bootswatch themes, custom colors, typography, brand.yml integration, Bootstrap Sass variables, custom Sass/CSS rules, dark mode and color modes, dynamic theme switching, real-time theming, theme inspection, or making R plots match the app theme with thematic.

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.