agent-auto-kit

Plugin agent-auto-kit from marketplace agent-auto-kit-marketplace in slenderman2511/AgentAutoKit. Commands read from the repo's manifest — 11 days ago.

Skills (35)

• accessibility

  Apply web accessibility (a11y) best practices to WCAG 2.2 AA when writing, reviewing, or fixing any UI, component, page, or form. Covers semantic HTML & landmarks, correct ARIA (name/role/value, and the "no ARIA is better than bad ARIA" rule), keyboard operability & visible focus, focus management f

• add-bug-to-github

  File a bug the user already knows about as ONE GitHub issue in the project's AI work queue — labeled with the queue label (default `ai`) + `bug` + `bug:<kind>` (data/ui/logic/integration/perf) + `severity:<level>` + optional scope labels, all as named in CLAUDE.md "## Issue tracking" — after checkin

• archify

  Create validated architecture, workflow, sequence, data-flow, lifecycle/state, ERD, class, tree, timeline, and waterfall diagrams as explorable standalone HTML with dark/light themes, motion and image/SVG/WebM export. Accept requirements or Mermaid; inspect repository evidence for code-backed diagra

• vercel-composition-patterns

  React composition patterns that scale. Use when refactoring components with

• context-checkpoint

  Keep a long Claude Code session healthy — persist durable state, prompt a compaction at a safe idle point, and run the save-memory → update-SOT → clear → resume ritual when a session ends or the work switches to a new feature/task. Use when context is getting heavy (~half full) and you've just reach

• conventions

  Project conventions for AgentAutoKit-managed repos — npm, TypeScript, Vitest, Vercel. Use when writing or reviewing code in this repo.

• e2e-flow

  Use when running, debugging, or authoring end-to-end user-journey Playwright specs in this repo (tests/*.spec.ts) — e.g. registration/payment flows like pwc2026-flow or team-format-flow, adding a new full-journey spec, wiring its env/seeding prerequisites, or diagnosing a failed flow run (test-resul

• e2e-visual-proof

  Use when an end-to-end run needs PROOF with pictures — screenshot every meaningful step of a real user journey (desktop, phone, or both), fix visible UI bugs found in the shots, and publish ONE self-contained review page (step gallery + API evidence table + honest notes) for the PR/issue. Runs as th

• firebase-best-practices

  Apply Firebase best practices when writing, reviewing, correcting, or hardening Firebase code and config — Firestore/Storage security rules, composite & single-field indexes (firestore.indexes.json), Firebase Auth, role/permission (RBAC) design, Cloud Functions, Realtime Database, and Remote Config.

• firestore-config-edit

  Edit tenant/event CONFIG directly in Firestore (theme, brand, feature flags, nav text) and bust the app cache so the change shows up. Use whenever flipping something on a tenant or event doc outside the admin UI — brand.themeClass, showNavText, landing config, featureFlags. Explains the 1-hour tenan

• fix-from-github

  Drain the AI work queue — pick the next unclaimed open issue carrying the project's queue label (default `ai`, from CLAUDE.md "## Issue tracking"; bugs from add-bug-to-github ordered by severity, then features), claim it with an advisory lock comment, investigate, present a per-ticket PLAN GATE for

• frontend-design

  Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautify

• git-workflow

  Apply Git best practices when syncing, branching, merging, or resolving conflicts, and when running multiple agents in parallel with git worktrees. Use when asked to "push/pull/sync", "merge a branch", "rebase", "resolve a merge conflict", "set up worktrees for parallel agents", "run several agents

• github-issue-flow

  Turn any change request (fix, update, new feature) into a tracked GitHub issue BEFORE any code, then drive it through the kit's pipeline — classify it (feature / bug / chore + optional scope), size it (S/M/L), open a well-labeled issue, link the worktree branch and draft PR to it, and close it after

• grill-me

  Use BEFORE planning a feature bigger than a day, or whenever a spec/plan still has open decisions — interview the user in ROUNDS, each round asking every decision whose prerequisites are already answered, each question carrying Claude's recommended answer so the user can reply "Q1 yes, Q2 no, Q3 rol

• hallmark

  Anti-AI-slop design skill for greenfield pages, audits, redesigns, and design extraction from URLs or screenshots. Use when the user asks to build a new app or landing page, wants to redesign something, invokes Hallmark by name, or uses audit/redesign/study.

• i18n-best-practices

  Apply i18n/multi-language best practices when writing, reviewing, or hardening any user-facing feature in a bilingual (EN/VI) or multi-locale project — catching hardcoded strings, enforcing translation-key conventions, keeping locale files in parity (no missing/orphan keys across en & vi), correct i

• nodejs-backend-patterns

  Build production-ready Node.js backend services with Express/Fastify, implementing middleware patterns, error handling, authentication, database integration, and API design best practices. Use when creating Node.js servers, REST APIs, GraphQL backends, or microservices architectures.

• nodejs-best-practices

  Node.js development principles and decision-making. Framework selection, async patterns, security, and architecture. Teaches thinking, not copying.

• payment-integration

  Apply payment best practices when integrating, reviewing, correcting, or hardening online payment flows across gateways — Stripe, Apple Pay, Google Pay, 9Pay, and SePay. Use when asked to "add online payments", "integrate Stripe/9Pay/SePay", "accept Apple Pay / Google Pay", "verify a payment webhook

• playwright-best-practices

  Use when writing Playwright tests, fixing flaky tests, debugging failures, implementing Page Object Model, configuring CI/CD, optimizing performance, mocking APIs, handling authentication or OAuth, testing accessibility (axe-core), file uploads/downloads, date/time mocking, WebSockets, geolocation,

• project-docs

  Set up and maintain the 8 core project docs an AI agent needs before writing code — PRD, DESIGN_SYSTEM, ARCHITECTURE, DATABASE, SECURITY, CODE_STYLE, TESTING, AGENTS — as numbered-section markdown in docs/, plus a one-file HTML viewer with a completion sidebar. Use when starting a new project or fea

• vercel-react-best-practices

  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,

• responsive-design

  Apply responsive-design best practices so UI renders well across desktop, tablet, and mobile when writing, reviewing, or fixing any layout, page, or component. Covers mobile-first breakpoints and fluid layouts (grid/flex, container queries), responsive images & fluid typography (srcset/sizes, next/i

• roster-import

  Import a roster from Excel (.xlsx) into Firestore as event entries for this pickleball tournament app. Use when the user hands over a spreadsheet of athletes/delegations and asks to seed entries, or references files under docs/tenants/**. Encodes the 6 data-seeding rules, the exact Entry/Participant

• seo

  Optimize for search engine visibility and ranking. Use when asked to "improve SEO", "optimize for search", "fix meta tags", "add structured data", "sitemap optimization", or "search engine optimization".

• stripe-best-practices

  Guides Stripe integration decisions — API selection (Checkout Sessions vs PaymentIntents), Connect platform setup (Accounts v2, controller properties), billing/subscriptions, Treasury financial accounts, integration surfaces (Checkout, Payment Element), migrating from deprecated Stripe APIs, and sec

• tailwind-css-patterns

  Provides comprehensive Tailwind CSS utility-first styling patterns including responsive design, layout utilities, flexbox, grid, spacing, typography, colors, and modern CSS best practices. Use when styling React/Vue/Svelte components, building responsive layouts, implementing design systems, or opti

• test-guard

  Keep a test suite honest and CI cheap. Use when creating tests for a change (`create`), auditing an existing suite for tests that cannot fail, skipped/commented tests and redundant cases (`audit`), or cutting GitHub Actions minutes and wall-time with measured, blast-radius test lanes (`optimize`, op

• typescript-advanced-types

  Master TypeScript's advanced type system including generics, conditional types, mapped types, template literals, and utility types for building type-safe applications. Use when implementing complex type logic, creating reusable type utilities, or ensuring compile-time type safety in TypeScript proje

• ui-prototype

  Build a throwaway UX PROTOTYPE for a NEW feature/screen before any production code — one self-contained HTML file styled with the project's REAL design tokens (CLAUDE.md "## Design system"), realistic domain mock data, and an interactive state machine that demos the UX decisions; published as a Clau

• ui-verify

  Verify UI changes AFTER implementation — catch broken layouts at mobile/tablet/desktop breakpoints, compute WCAG color contrast under every theme/variant the component ships in, and enforce the project's design-system rules (no hardcoded colors, bg↔text pairing, the prohibitions listed in CLAUDE.md

• upgrade-stripe

  Guide for upgrading Stripe API versions and SDKs

• voice-clone

  Clone a consenting person's voice locally — speech (text → their voice, viXTTS/XTTS) or singing (swap the vocal of an existing song to their voice, Seed-VC) — and score the result without listening. Use when the user asks to "clone a voice", "đổi giọng", "lồng tiếng bằng giọng của X", make a voiceov

• worktree-dev

  Set up a git worktree that can actually run the dev server and scripts — for any feature, fix or agent task that needs an isolated working copy (the worktree-first rule in .claude/rules/workflow.md), for running a second dev server, or when several Claude sessions share one checkout and collide. Als

Slash commands (3)

• /agent-auto-kit:init-kit

  Kick off the AgentAutoKit workflow for a feature or bug — size it (S/M/L), route it to the specialist agents, and run the review gate

• /agent-auto-kit:kit-stats

  Summarize AgentAutoKit run metrics into a routing scorecard

• /agent-auto-kit:kit-tune

  Propose (or apply) model-tier changes from the routing scorecard

Subagents (11)

• api-data-reviewer

  Use this agent to review API routes and data-model changes as a CONTRACT shared by every client (web, mobile, other services) — backward compatibility / versioning, payload shape that works for all clients, schema design (scoping fields, optional new fields, indexes, document growth), and query/read

• arch-advisor

  Use this agent when making architecture or design decisions — evaluating approaches for a new feature, designing data models or schema changes, planning multi-file refactors, assessing security, payment or data-isolation tradeoffs, or reviewing a plan before implementation. Use proactively before an

• code-reviewer

  Use this agent to review code changes against this project's rules (CLAUDE.md and .claude/rules/) plus correctness, UI states, TypeScript conventions, performance and test coverage. Run ONCE per batch of changes (before opening a PR), in parallel with security-auditor when that applies — not after e

• code-scout

  Use this agent when you need read-only codebase exploration — finding files, locating functions/components/hooks, listing call sites, tracing where a symbol is used, checking config values (package.json, tsconfig, framework config, env usage), or summarizing how an existing feature works. Use proact

• deep-debugger

  Use this agent when facing hard debugging or complex multi-file refactors — bugs with unknown root cause, race conditions, async/state bugs, complex generics, webhook or payment failures, data mismatches, build failures with unclear origin, or refactors that touch many files with behavior-preservati

• github-workflow

  Git workflow agent for complex git operations ONLY — multi-branch work, rebases, conflict resolution — following this project's conventions (feature branches off the integration branch, Conventional Commits, PRs always target the integration branch). A single commit + PR is done by the main session

• implementer

  Use this agent when implementing routine, well-scoped code changes — small features, bug fixes with a known cause, refactors that follow existing patterns, UI tweaks, adding an API route or query that mirrors an existing one, i18n strings, or wiring props/handlers. Use proactively for any change tou

• orchestrator

  Coordinator that sizes a task (S/M/L), routes it to the specialist agents, and enforces the review gate. The preferred way to orchestrate is the main session running /init-kit (it can stop and ask the user); use this agent only for a fully delegated run. Never writes code itself.

• security-auditor

  Use this agent when reviewing security-sensitive code — payment webhooks and callbacks (signature verification), security rules and access policies, data isolation (per-tenant or per-user scoping), RBAC and API route auth, secrets handling, or input validation at trust boundaries. Run ONCE per PR on

• spec-reviewer

  Review a design spec in docs/superpowers/specs/ BEFORE the implementation plan is written (L-tier tasks only). Call this agent immediately after finishing a spec file and before writing the first line of the plan. It returns a list of gaps with file:line; read-only, it never edits the spec. Do not s

• test-writer

  Use this agent when writing or updating tests — Vitest unit tests for business logic, Playwright e2e specs, test factories, or fixing failing/flaky tests after a code change. Use proactively after implementing logic that lacks coverage.

Hooks (14)

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/turn-guard.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/protect-files.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/test-guard.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/git-guard.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/review-gate.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/test-budget.sh

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/agent-routing.sh

• PostToolUse

  "${CLAUDE_PLUGIN_ROOT}"/hooks/review-gate.sh

• SubagentStop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/metrics-subagent.sh

• Stop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/verify.sh

• Stop

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-state.sh

• SessionStart

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-state.sh

• UserPromptSubmit

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-state.sh

• SessionEnd

  "${CLAUDE_PLUGIN_ROOT}"/hooks/session-state.sh

Similar plugins