aac-skills

Create, review, or audit an Active Alarm Company (AAC) customer contract package — the Schedule of Equipment and Services, the master agreement, and the Rider for Additional Locations. Use whenever a rep or Sales Admin asks to create, draft, build, or "put together" a contract, schedule, or paperwork for a named job or customer; when a "please create contract" request is forwarded; when a job folder path is given and the ask is paperwork; when a package needs rebuilding after a fact changes (price, purchase vs. financed, designation, scope); or when an existing package needs reviewing, verifying, or sweeping for defects across the jobs drive. Also use for questions about AAC clarifications, exclusions, RMR names and prices, SOW wording, the $5,000 deposit rule, permit responsibility, or the Schedule-to-Master mapping. Prefer this over generic document generation or contract review for anything touching an AAC package, even if the words "schedule" or "package" are not used.

Durable owner-grade access to the AAC Google stack (GCP project, service account, clasp token, Apps Script projects). Load before any Sheets, Drive, Gmail or Apps Script work in an AAC project, when a Google API call fails on auth or scope, when clasp needs re-authenticating, or before claiming you lack Google access.

AAC-WR-001, the controlled copy of AAC's house writing and document standard. Load before drafting, formatting or reviewing any AAC deliverable: email, Teams message, memo, letter, report, SOP, proposal, scope, Word document, or table. Also load for a performance review or its audit.

Audit an AAC One Page performance review draft from a reviewing manager and produce the skip-level's email back. Use whenever Dan uploads or points at a performance review draft, revision, or self-appraisal, asks to "audit," "check," or "gate" a review, or asks for the rejection or coaching email for one.

Check an AAC One Page performance review draft against the house standards before it goes to the skip-level. Use whenever the user is writing, revising, or about to send a performance review for one of their directs.

Create or rewrite an Active Alarm Company (AAC) procedure or work instruction in AAC's house format and writing standards, delivered as a styled Word (.docx). Two tiers: a full Procedure (cross-role process) and a lean Work Instruction (one person, one task, usually one tool). Use whenever the user asks to write, draft, standardize, or recast an SOP, "standard operating procedure," "work instruction," or "WI" - or to put an existing process, checklist, or rough document "into our SOP format." Also use when documenting any repeatable AAC process or task (service calls, monitoring termination, billing, payroll, the Service-to-Sales handoff, estimating, month-end close, or a tool how-to such as System Surveyor or Zoho). Trigger even without the words "template," "SOP," or "work instruction" - "write up how we handle X," "document the X process," or "turn this into a procedure/checklist" all apply. Prefer this over generic process-doc/runbook output for any AAC procedure or work instruction.

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. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.

Ask which skill or flow fits your situation. A router over the skills in this repo.

Ultra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

Audit or trim a CLAUDE.md, AGENTS.md or other always-loaded instructions file against the concision paradigm ("would removing this line cause a mistake?"). Use when asked to lint, audit, trim, shrink or review a CLAUDE.md, when a rulebook feels ignored or bloated, or before adding a new section to one. Bundles the deterministic linter; the trim pass is the model's job.

Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/PRD asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use when the user wants to review a branch, a PR, work-in-progress changes, or asks to "review since X".

Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

Repo-wide prose-vs-reality audit. Verifies every factual claim in the repo's documents (README, CLAUDE.md, ADRs, PRDs, runbooks, trace docs, state logs, agent memory) against primary sources — code, git, the tracker, live state — fixes what is stale, hunts the scripts that generate drift, and reports root causes. Use when docs contradict each other or claim work states wrongly ("X is already implemented" when it isn't, or the reverse).

Full holistic pass over every memory file — merge duplicates, fix stale facts, prune the index. Never a delta sweep.

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

Deploy, promote, run and diagnose an AAC Apps Script project without clasp, through the gas package in claude-dotfiles (a vendored SelfDeploy.js, deploy/* refs on GitHub, one time trigger). Use when a repo's clasp credential died again, when asked to "move this repo off clasp", "adopt gas", "deploy the script", "promote PROD", "run function on the script", or to read why a gas/deploy status is red.

A relentless interview to sharpen a plan or design.

Walk every ready-for-human ticket one at a time — grill for the ruling, land it as a comment, relabel or close.

A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.

Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

Compact the current conversation into a handoff document for another agent to pick up.

Implement a piece of work based on a spec or set of tickets.

Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

Weekly repo hygiene — fix doc-vs-reality drift, then tidy memory. Run by hand.

Automate browser interactions, test web pages and work with Playwright tests.

Bolt the production organization harness onto any repo — triage labels, issue forms, generated DASHBOARD.md + CI refresh, pre-commit test gate, ADR status lines, live tracker-drift audit, Projects board. Use when the user says "harness this repo", "set up the project harness", "make this repo organized like aac-cockpit", "upgrade the harness", or spins up a new project. Idempotent — safe to re-run, and carries a version marker so an existing install can be upgraded.

Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.

Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.

Use when you need to resolve an in-progress git merge/rebase conflict.

Engine behind the session gate — check.js runs the git/clasp/test/ticket checks for any repo. The hooks call it; /session-start and /session-end re-print its report.

Land the session — commit, push, merge, file what was promised, re-run the end check — and end on the archive line. The checks already run as a hook when a turn reads as wrapping up; this is the sequence that clears them.

Re-print the start-of-session checks for a git project — the remote, uncommitted work, the deploy credential, tests, tickets. Use when that report has scrolled out of context, or with --refresh to re-run it after the tree has moved.

Configure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.

Decide whether to adopt a new paid software tool, with the analysis scaled to the size of the decision. Use when someone proposes buying or subscribing to software ("should we get [tool]"), wants a tool evaluated, or is comparing tools — from a low-cost utility to a multi-year platform.

Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

Teach the user a new skill or concept, within this workspace.

Parallel ticket runner: scout, pinned implementer per ticket, blind refuting verifier, PR on pass, discovery collection. Drives open ready-for-agent tickets to verified PRs in parallel; also runs probe tickets (evidence in a comment) and ready-for-human tickets (verify what a container can, hand the rest to the owner). One script served by this plugin, invoked via the Workflow tool with `scriptPath` from local and cloud sessions alike; the script picks between the `gh` CLI and the GitHub MCP tools at run time. Use when the user asks to run the ticket fleet, clear a wave of `ready-for-agent` tickets, or invoke the fleet from an orchestrator worker cycle.

Sweep every open ticket in the current repo against the speed-over-robustness rule — park belt-and-suspenders work in the Maybe Someday milestone, close only what is moot, post one digest. Runs as step 0 of the weekly /maintain-repo; a standalone weekly Routine is not created yet (issue 664).

Turn the current conversation into a spec and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker — edges as text in one file per ticket locally, or native blocking links on a real tracker.

Triage Dan's Todoist work projects. Use when Dan asks to triage tasks, clear the backlog, run the daily or Friday pass, or decide what to delegate.

Move issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.

Rebuild and republish the aac-skills plugin (the single package built from the repo's aac-skills/ tree) when the session-end sweep reports drift. Merging the rebuild is the primary channel; zip upload is the fallback for claude.ai Skills pages. Use when the session-end cloud-skills sweep reports drift or no recorded upload, when the user says the cloud sessions are missing a skill, or after adding or editing a skill that should reach claude.ai/code and Cowork.

React composition patterns that scale. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or designing reusable APIs. Triggers on tasks involving compound components, render props, context providers, or component architecture. Includes React 19 API changes.

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

React Native and Expo best practices for building performant mobile apps. Use when building React Native components, optimizing list performance, implementing animations, or working with native modules. Triggers on tasks involving React Native, Expo, mobile performance, or native platform APIs.

Guide for implementing smooth, native-feeling animations using React's View Transition API (`ViewTransition` component, `addTransitionType`, and CSS view transition pseudo-elements). Use this skill whenever the user wants to add page transitions, animate route changes, create shared element animations, animate enter/exit of components, animate list reorder, implement directional (forward/back) navigation animations, or integrate view transitions in Next.js. Also use when the user mentions view transitions, `startViewTransition`, `ViewTransition`, transition types, or asks about animating between UI states in React without third-party animation libraries.

Plan a huge chunk of work — more than one agent session can hold — as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.