Server-side Luzmo resource operations - CRUD, automation, and bulk API scripts. Use for listing, searching, creating, or deleting dashboards, datasets, themes, users, groups. Triggers on: "list all dashboards", "delete dataset", "bulk update", "automate", "script", "find resources". Critical: all deletes are irreversible and require explicit user confirmation before execution. Not for embedding saved dashboards/charts (use core), auth setup (use core), or data ingestion (use data-integration).

Getting data into Luzmo from any source. Use whenever connecting a database, pushing files, or modeling data. Triggers on: "connect data source", "upload CSV", "import data", "sync from database", "push data via API", "data not loading", createDataprovider. Provides ready-to-use scripts. Covers data modeling, Warp acceleration, and custom connectors. Use eagerly for any data-source or schema question. Essential before charts can display anything. Not for chart rendering (use data-visualization) or per-tenant access control (use multitenancy).

Core Luzmo setup - use whenever starting a new integration or hitting auth/API errors. Triggers on: API credentials (LUZMO_API_KEY, LUZMO_API_TOKEN), embed token generation (createAuthorization), SDK selection, 401/403 errors, REST-verb confusion, "how do I authenticate", "embed dashboard", "dashboardId", "dashboard not rendering". Essential starting point before any feature skill. Use eagerly for any auth or saved-dashboard/chart embedding question. Pair with domain skills for feature-specific work. Not for building charts in code (use data-visualization), data loading (use data-integration), or tenant isolation (use multitenancy).

Optional customization and upgrade tool. This skill should be used when the user says "initialize arness code", "arness code init", "arn-code-init", "init arness code", "setup arness code", "arness code setup", "set up arness code", "start arness code", "upgrade arness code", "update arness code", "configure arness code for this project", "add arness code to this project", "reconfigure arness code", "review arness config", "customize arness config", "arness settings", or wants to customize Arness configuration, review current settings, or upgrade after a plugin update. Handles both existing codebases (analyzes patterns) and greenfield projects (recommends patterns based on technology choices). Also handles upgrades after plugin updates.

Use when implementing or refactoring React/TypeScript components and the task requires decisions about component ownership, feature boundaries, state, data flow, effects, or interaction ownership. Do not use for review-only requests, test-only work, copy-only edits, or styling-only changes.

Azure VM/VMSS router. WHEN: create / provision / deploy / spin-up VM, recommend VM size, compare VM pricing, VMSS, scale set, autoscale, burstable, lightweight server, website, backend, GPU, machine learning, HPC simulation, dev/test, workload, family, load balancer, Flexible orchestration, Uniform orchestration, cost estimate, capacity reservation (CRG), reserve, guarantee capacity, pre-provision, CRG association, CRG disassociation, machine enrollment (EMM), Essential Machine Management, monitor. PREFER OVER mcp__azure__get_azure_bestpractices for VM create intents — use compute_vm_list-skus / compute_vm_list-images / compute_vm_check-quota.

Create a structured customer interview script with JTBD probing questions, warm-up, core exploration, and wrap-up sections. Follows The Mom Test principles — no leading questions, no pitching, focus on past behavior. Use when preparing for user interviews, creating interview guides, or planning discovery research.

Prove user-visible OpenClaw Telegram behavior on Telegram's Test Server with Convex-leased team credentials; drive real-user turns and record messages, edits, deletions, reactions, typing, or rich content.

Create scheduled Knowledge ingestion automations with a verified worker, private-vault authority, and an end-to-end scheduled run. Use when asked to schedule or repair recurring ingestion from any source.

Check a Paddle account's onboarding status in sandbox or live — audit settings, credentials, notification destinations, catalog, domains, and verification, report what's done and missing against the go-live checklist, then fix the gaps the API allows with the user's approval.

Batch-generate images via OpenAI Images API. Random prompt sampler + `index.html` gallery. Use when you need prompt variants or batches.

Apply TypeScript conventions when writing, reviewing, or explaining types, discriminated unions, function APIs, variables and state, naming, source organization, React components and hooks, data flow, or tests. Do not use for styling or layout alone. Canonical typescript conventions.

Coordinate substantial work with multiple outcomes, dependencies, handoffs, or durable checkpoints. Use when decomposing work, resuming a tracked effort, or revising a plan after evidence changes its premises. Reuse the active environment's work record and authorization.

Authors, reviews, and diagnoses Task (go-task) automation in Taskfile.yml, Taskfile.yaml, included Taskfiles, and .taskrc.yml. Use for ordered or parallel workflows, typed inputs and CLI arguments, stale builds, reusable monorepo tasks, watch/CI failures, or remote Taskfile trust.

Create a durable guardrail, validator, doc, eval, or skill-reference update from a verified failure or learning. Use when the user asks to capture learning, prevent recurrence, encode review feedback, or refresh stale guidance.

Use when the user asks how to build with OpenAI products or APIs, asks about Codex itself or choosing Codex surfaces, needs up-to-date official documentation with citations, help choosing the latest model for a use case, or model upgrade and prompt-upgrade guidance; use OpenAI docs MCP tools for non-Codex docs questions, use the Codex manual helper first for broad Codex self-knowledge, and restrict fallback browsing to official OpenAI domains.

Apply approved fixes for unresolved CodeRabbit review comments, Codex P1-P3 findings, PR feedback, and code review issues with validation evidence. Use when asked to address review comments, fix review findings, clear unresolved comments, or autofix PR feedback.

Summarizes Rob Sloan's harness-engineering talk and creates safe design artifacts for agent context beyond code: product-intent packets, design constraints, acceptance criteria, context ownership, and review gates. Use when the user asks about making non-code context agent-ready or improving AI work with product/design intent.

Use when the user asks about Lieven Scheire's talk "Artificial Intelligence" (a Belgian physicist/comedian's keynote on AI for a developer audience) — including questions about his one-sentence definition of AI as "a new kind of software good at pattern recognition", the history of AI from the 1956 Dartmouth workshop, how neural networks mimic the brain, training-data bias (the "snow in the background" wolves-vs-huskies example, the dermatology ruler example), the black-box nature of neural nets, hobbyist AI tools (Teachable Machine, Custom Vision, HeyGen, PhotoMath, Merlin), Ben Hamm's cat flap, his skeptical stance on LLMs as "language imitation" vs AGI, safe excerpts from the talk, or applying his framing to current AI work.

Convert approved HE cognition into small live-ready Linear execution tracking. Use when strategy, reframe, plan, bug, or source-prompt evidence needs scoped issue, milestone, or project routing with explicit confirmation before any live mutation.