Write, review, and edit documentation files with consistent structure, tone, and technical accuracy. Use when creating docs, reviewing markdown files, writing READMEs, updating `/docs` directories, or when user says "write documentation", "review this doc", "improve this README", "create a guide", or "edit markdown". Do NOT use for code comments, inline JSDoc, or API reference generation.

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

Deploy AI/ML models on OpenShift AI using KServe with vLLM, NVIDIA NIM, or Caikit+TGIS runtimes. Use when: - "Deploy Llama 3 on my cluster" - "Set up a vLLM inference endpoint" - "Deploy a model with NIM" - "Create an InferenceService for Granite" - "I need to serve a model on OpenShift AI" Handles runtime selection, GPU validation, InferenceService CR creation, and rollout monitoring. NOT for NIM platform setup (use /nim-setup first). NOT for custom runtime creation (use /serving-runtime-config).

Extracts an SMV (NuSMV/nuXmv) finite-state model from code or state-machine descriptions, for CTL/LTL model checking of reactive systems. Use when verifying hardware-adjacent or embedded logic, when the state space is naturally finite and small, or when CTL branching-time properties are needed.

Matches code against Project CodeGuard's catalog of known-dangerous patterns — banned C functions, weak crypto primitives, hardcoded credentials, deprecated APIs. Use when grepping for low-hanging security fruit, when enforcing a ban-list in CI, or when the user asks to check for known-bad patterns.

Scans source code for security vulnerabilities by applying Project CodeGuard rules — injection, unsafe deserialization, XSS, path traversal, broken access control. Use when performing a security audit, when reviewing a PR that touches request handlers or database queries, when the user asks for a vulnerability scan, or when wiring security checks into CI.

Query, audit, and optimize Google Ads campaigns. Use an attached browser or the Google Ads API for campaign, keyword, budget, conversion-tracking, and wasted-spend analysis, or when the user explicitly requests a reviewed account change. Defaults to read-only reporting; account mutations require a bounded preview and action-time approval.

This skill should be used when the user says "static prototype teams", "arn static prototype teams", "team static prototype", "debate static prototype", "collaborative visual review", "static prototype with debate", "team-based visual review", "visual debate", "review visuals as a team", or wants to create a static component showcase and validate it through iterative expert debate cycles where product strategist and UX specialist discuss their scores and findings before producing a combined review, with per-criterion scoring, an independent judge verdict, and versioned output. Supports Agent Teams for parallel debate or sequential simulation as fallback. For standard lower-of-two-scores visual review, use /arn-spark-static-prototype instead.

This skill should be used when the user says "spike", "arn spike", "validate risks", "technical validation", "proof of concept", "validate architecture", "risk spike", "test this risk", "will this work", "technical spike", "validate the stack", or wants to validate critical technical risks from the architecture vision by creating minimal proof-of-concept code and testing whether the chosen technologies work as expected.

This skill should be used when the user says "prototype lock", "lock prototype", "arn prototype lock", "freeze prototype", "preserve prototype", "snapshot prototype", "protect prototype", "archive prototype", "save the prototype", "don't overwrite the prototype", "lock the design", "freeze the design", or wants to create a frozen snapshot of the validated prototype before development begins, preventing production code from overwriting the validated reference artifact.

This skill should be used when the user says "infra change spec", "infrastructure change", "specify infrastructure change", "change spec", "arn infra spec", "describe infra change", "what infrastructure needs to change", "infra spec", "spec this infra change", "write infra spec", "infrastructure change specification", "create infra change spec", "upgrade to pipeline", "convert to pipeline", or wants to iteratively develop an infrastructure change idea into a well-formed specification through guided conversation, or wants to upgrade existing interactive IaC artifacts into the structured change pipeline.

Use when the user wants to build a Python Kafka producer or consumer, add Schema Registry to existing Python code, migrate from raw JSON to schema-backed serialization, or scaffold a confluent-kafka-python project for Confluent Cloud, local Docker, or WarpStream. Also use when user wants to optimize Python Kafka client configuration for WarpStream.

Use browser-use to perform Google searches, open results, and extract key information from live pages. Use when the user asks to "search Google", "look this up on Google", or needs current web results via a real browser session (often to avoid bot blocks).

Red-team a PRD, roadmap, or strategy by attacking its load-bearing assumptions before reality does. Steelmans then attacks each claim, ranks failure modes by impact × likelihood × cheapness-to-test, and returns the cheapest test and kill criteria for each. Use when stress-testing a plan, pressure-testing a strategy, challenging assumptions, or preparing a doc for executive review.

This skill should be used when the user asks to "get a second opinion", "ask codex", "what does GPT think", "compare with codex", "run codex", "2nd opinion", "second opinion on this code", or wants an independent analysis from OpenAI Codex CLI. Also triggered by the /codex-2nd-opinion command.

Trust Wallet open-source libraries — Wallet Core (HD wallets, address derivation, tx signing in Swift/Kotlin/TypeScript/Go for 140+ chains), Web3 Provider (dApp connection for Ethereum/Solana/Cosmos/Bitcoin/Aptos/TON/Tron), deep linking, browser extension integration, WalletConnect, token assets repository, and Barz ERC-4337 smart wallet. Use when working with trustwallet/wallet-core, @trustwallet/wallet-core, trust-web3-provider, Trust Wallet deep links, token logos/metadata from trustwallet/assets, or Barz account abstraction.

Trust Wallet API for crypto data — token search, prices, trending tokens, swap quotes, market data, security checks, address validation, asset info, and coin status across 100+ blockchains. Use whenever the user asks about crypto prices, token info, swap rates, market cap, trending coins, token risk, honeypot detection, address validation, or wants to call the Trust Wallet / tws.trustwallet.com API directly. Covers HMAC-SHA256 authentication, supported chains, and all REST endpoints.

Generate Excalidraw diagrams from text content. Supports three output modes - Obsidian (.md), Standard (.excalidraw), and Animated (.excalidraw with animation order). Triggers on "Excalidraw", "画图", "流程图", "思维导图", "可视化", "diagram", "标准Excalidraw", "standard excalidraw", "Excalidraw动画", "动画图", "animate".

Search Cupix Watch (Kibana/Elasticsearch) application logs at watch.cupix.com. Use when the user asks to search logs, find errors, debug processing issues, or investigate service behavior. Triggers on keywords like "log", "watch", "kibana", "error log", service names (skat, pano, api, worker, vista), or mentions of cupix processing pipelines.

Author, edit, or apply a Descope FGA schema using the ReBAC/ABAC DSL. Use this skill whenever the user asks to create a new FGA schema, modify an existing one, add types/relations/permissions/conditions, review an authorization model, or apply schema changes to a Descope project. Trigger even if the user says things like "set up authorization", "define roles and permissions", "add team-based access", "make this endpoint check FGA", or "update my authz model" — these almost always mean an FGA schema change.