Skills cheatsheet

AgentKit Skills — Engineer & Marketing

Skill catalog by workflow group, with subcommands, args, and flags. Claude Code & Cursor: /ak:<skill> · Codex: $ak:<skill>.

Engineer = core + engineer skills. Marketing owns domain skills. Do not assume 1:1 with /ck:*. Syntax from kit argument-hint and SKILL.md.

108
Engineer
90
Marketing
/ak: · $ak:
Syntax
Chip legend
archive
Subcommand
Skill branch: /ak:plan archive
[task]
Argument
Positional / placeholder args
--fast
Flag
Dash options like --flag
hint
argument-hint
Full syntax from SKILL.md

Beta Preview — what is new in beta

Skills that are new in beta and not yet in the stable kit. They also get a BETA badge in the groups below.

/ak:page-builder new

$ak:page-builder

Integrate or extend a project page builder, then operate drafts, previews, and publishing. Not one-off page design.

[setup|update|operate][project-path][request]

Start Here (Essentials)

Core path: Level → Ask → Brainstorm → Plan → Preview → Execute

9

$ak:coding-level

/ak:coding-level [0-5]

Set coding experience level for tailored output. Use for adjusting explanation depth, code complexity, and response format to user expertise.

Open visual guide →
[0-5]

$ak:ask

/ak:ask [technical-question] [--yagni]

Answer technical and architectural questions with expert analysis. Use for design decisions, best practices evaluation, solution comparison.

Open visual guide →
[question]
--yagni

$ak:brainstorm

/ak:brainstorm [topic or problem] [--advice] [--html] [--report] [--ultra] [--yagni] [--no-antv|--no-diagram-design|--no-editorial-visuals]

Turn unclear intent into an accepted outcome and compare viable approaches before delivery.

Open visual guide →
[topic or problem]
--advice--html--no-antv--no-diagram-design--no-editorial-visuals--report--ultra--yagni
4 /ak:plan COMMON

$ak:plan

/ak:plan [task] [--fast|--hard|--deep|--parallel|--two|--debate|--ultra] [--tdd|--no-tasks] [--html] [--github] [--wiki] [--advice] [--yagni] [--skip-journal] [--no-antv|--no-diagram-design|--no-editorial-visuals] OR [archive|red-team|validate]

Plan implementations, design architectures, create technical roadmaps with detailed phases. Use for feature planning, system design, solution architecture, implementation strategy, phase documentation, editorial self-contained HTML plan artifacts with --html, and AgentWiki publishing with --wiki.

Open visual guide →
archivered-teamvalidate
[task]
--advice--debate--deep--fast--github--hard--html--no-antv--no-diagram-design--no-editorial-visuals--no-tasks--parallel--skip-journal--tdd--two--ultra--wiki--yagni

$ak:preview

/ak:preview [path] OR [--html] --explain|--slides|--diagram|--ascii [topic] OR --html --diff|--plan-review|--recap [--no-antv|--no-diagram-design|--no-editorial-visuals]

View files or generate visual explanations, slides, and diagrams. Use for code walkthroughs, architecture visualization, HTML/Markdown presentations.

Open visual guide →
[path][topic]
--ascii--diagram--diff--explain--html--no-antv--no-diagram-design--no-editorial-visuals--plan-review--recap--slides
6 /ak:cook COMMON

$ak:cook

/ak:cook [task|plan-path] [--interactive|--fast|--parallel|--auto|--no-test] [--tdd] [--advice] [--yagni] [--skip-journal]

Implement features, plans, and fixes with structured workflow. Use for feature development, plan execution, code implementation pipelines.

Open visual guide →
[task|plan-path]
--advice--auto--fast--interactive--no-test--parallel--skip-journal--tdd--yagni

$ak:advise

/ak:advise [prompt-or-url] [--html] [--md] [--wiki] [--github] [--agent] [--ultra] [--yagni] [--no-antv|--no-diagram-design|--no-editorial-visuals]

Interview-driven advisory skill. Analyzes a prompt or URL (GitHub issue, spec, doc), scouts the codebase, interviews the user one question at a time to reframe the problem into exact requirements and goals, then delivers honest advice: what to do, what to avoid, better alternatives, benefits and trade-offs. Use when the user asks for advice, a second opinion, a sanity check, requirement reframing before committing to a direction, or wants an existing plan or design pressure-tested (grilled) one question at a time.

Open visual guide →
[prompt-or-url]
--agent--github--html--md--no-antv--no-diagram-design--no-editorial-visuals--ultra--wiki--yagni

$ak:vibe

/ak:vibe [--ship] [--beta] [--both] [--advice] [--ultra] <github-issue-url | feature request>

Run the full vibe pipeline from request intake to PR readiness, with optional merge and post-merge CI convergence. Orchestrates worktree, plan, cook/fix, code-review, ship, and review-pr. Supports dual-stage beta-then-stable ships via --both and kongming advisory supervision via --advice. Use for GitHub issues, feature requests, bug fixes, or autonomous ship runs.

Open visual guide →
<github-issue-url | feature request>
--advice--beta--both--ship--ultra

$ak:bro

/ak:bro [last message | topic to simplify]

Restate the assistant's last message in simpler, shorter, jargon-free language. Use when the user says ak:bro, simplify that, say it plainly, or explain it like a human.

Open visual guide →
[last message | topic to simplify]

Plan & Research

Brainstorming, planning, advisory, and codebase exploration

12

$ak:autoresearch

/ak:autoresearch <goal or hypothesis> [--iterations N] [--metric <metric>]

Route bounded, goal-directed iteration to the AgentKit skill that owns the desired outcome.

Open visual guide →
<goal or hypothesis>
--iterations--metric

$ak:codex-goal

/ak:codex-goal <objective | goal draft>

Guide long-running Codex goal work with a verifiable stop condition. Use when users mention /goal, goal mode, durable objectives, or autonomous multi-turn Codex runs.

Open visual guide →
objective

$ak:fable-thinking

/ak:fable-thinking [task or question to reason through]

Reasoning protocol distilled from Claude Fable 5. Makes any model reason like Fable — evidence-grounded claims, multi-hypothesis diagnosis, concrete simulation, adversarial self-review, calibrated outcome-first delivery. Its never-skipped Floor check catches simple-looking trick questions models answer confidently wrong, and its Constraint Loop mechanically verifies hard output constraints (banned letters, exact counts, strict formats) that models otherwise rubber-stamp. Use for debugging, review, analysis, decisions, constrained writing, or any task where being right matters more than being fast.

Open visual guide →
[task or question to reason through]

$ak:goal-warmup

/ak:goal-warmup "<goal>" [--fast]

Outcome-locked preflight before long-running /goal or autonomous runs. Interview to a user-approved Outcome Contract, plan without silent scope drift, contract-preserving review, whole-plan preflight matrix, then Ready/Blocked/Decision handoff. Never auto-starts /goal. Use for goal warmup, goal prepare, long-running goal prep, outcome lock, execution readiness.

Open visual guide →
<goal>
--fast

$ak:issue-to-plan

/ak:issue-to-plan <github-issue-url | issue-number> [--repo owner/name] [--plan-ready-label <name>] [--decision-label <name>]

Turn a GitHub issue into an audited, validated implementation plan. Reads the issue, scouts the codebase, runs a hard brainstorm gate, and only then plans with mandatory --html --wiki, validate, and red-team, before pushing a plan branch and handing off on the issue. Use to convert a GitHub issue into a validated plan that is ready for plan audit.

Open visual guide →
<github-issue-url | issue-number>
--decision-label--plan-ready-label--repo

$ak:predict

/ak:predict <feature description or change proposal> [--files <glob>] [--chain reason|probe]

5 expert personas debate proposed changes before implementation. Catches architectural, security, performance, and UX issues early. Use before major features or risky changes.

Open visual guide →
<feature description or change proposal>
--chain--files

$ak:problem-solving

/ak:problem-solving [problem description] [--ultra]

Apply systematic problem-solving techniques when stuck. Use for complexity spirals, innovation blocks, recurring patterns, assumption constraints, simplification cascades, scale uncertainty.

Open visual guide →
[problem description]
--ultra

$ak:research

/ak:research [topic] [--ultra] [--yagni]

Research technical solutions, analyze architectures, gather requirements thoroughly. Use for technology evaluation, best practices research, solution design, scalability/security/maintainability analysis.

Open visual guide →
[topic]
--ultra--yagni

$ak:research-prompt

/ak:research-prompt <research topic | decision>

Draft a self-contained research brief for a human or AI researcher. Use when users ask for a research prompt, research brief, or a deep-research task to hand off.

Open visual guide →
decision

$ak:scout

/ak:scout [search-target] [ext] [--ultra]

Fast codebase scouting using native search, optional Explore agents, and user-permitted OpenCode probes. Use for file discovery, task context gathering, and scoped searches across directories.

Open visual guide →
[search-target][ext]
--ultra

$ak:sequential-thinking

/ak:sequential-thinking [problem to analyze step-by-step]

Apply step-by-step analysis for complex problems with revision capability. Use for multi-step reasoning, hypothesis verification, adaptive planning, problem decomposition, course correction.

Open visual guide →
[problem to analyze step-by-step]

$ak:sowat

/ak:sowat [recent-changes|issue|PR]

Analyze recently implemented work and related issues like a product owner. Use to identify high-impact next steps, challenge weak priorities, and explain what matters now.

Open visual guide →
[recent-changes|issue|PR]

Design & Frontend

UI/UX, frontend, diagrams, and visual tools

17

$ak:design

/ak:design [design-type] [context]

Design brand identity, logos, banners, posters, and visual assets. Use for brand systems, design tokens, corporate identity programs, event/editorial/marketing posters, and visual campaign assets. Not for UI code patterns.

Open visual guide →
[design-type][context]

$ak:diagram

/ak:diagram [input-file] [--format <svg|fragment|html>] [--preset <classic|signal-flow|blueprint|editorial>] [--theme <light|dark>] [--out <path>]

Unified diagram surface — Mermaid, editorial diagrams (24 base types across architecture, flow, storytelling, data-viz), and animated SVG connectors. Use when the user wants a static image (PNG/SVG), a self-contained editorial HTML page, or a short video (MP4 default, GIF via --gif) of a diagram at editorial quality. Deterministic byte-for-byte output.

Open visual guide →
[input-file]
--format--out--preset--theme

$ak:excalidraw

/ak:excalidraw [diagram description|path] [--export <png|svg>] [--live]

Create Excalidraw diagrams — architecture, data flow, workflows, system design. Use when user wants to visualize, diagram, draw architecture, show data flow, create flowcharts, map components, or export .excalidraw files to PNG/SVG. Supports two modes: live MCP canvas (real-time) or file-based JSON + Playwright rendering. Also supports zero-config codebase auto-diagramming — just say "diagram this repo" or "visualize the architecture".

Open visual guide →
[diagram description|path]
--export--live

$ak:frontend-design

/ak:frontend-design [prompt|image-path|component]

Create polished frontend interfaces from designs/screenshots/videos. Use for web components, 3D experiences, replicating UI designs, quick prototypes, immersive interfaces, avoiding AI slop.

Open visual guide →
[prompt|image-path|component]

$ak:frontend-development

/ak:frontend-development [component or feature]

Build React/TypeScript frontends with modern patterns. Use for components, Suspense, lazy loading, useSuspenseQuery, MUI v7 styling, TanStack Router, performance optimization.

Open visual guide →
[component or feature]

$ak:mermaidjs-v11

/ak:mermaidjs-v11 [diagram-type or description]

Create diagrams with Mermaid.js v11 syntax. Use for flowcharts, sequence diagrams, class diagrams, ER diagrams, Gantt charts, state diagrams, architecture diagrams, timelines, user journeys.

Open visual guide →
[diagram-type or description]

$ak:page-builder

/ak:page-builder [setup|update|operate] [project-path] [request]

Build, extend, or operate a dynamic page builder using a project's components and stack. Use for configurable resizable widgets, responsive touch editors, previews, publishing, and agent interfaces with a project operator skill. Not for one-off page design or a generic API wrapper.

Open visual guide →
setupupdateoperate
[project-path][request]

$ak:react-best-practices

/ak:react-best-practices [component or pattern]

Apply React and Next.js performance optimization patterns from Vercel Engineering. Use for component optimization, rendering performance, bundle analysis.

Open visual guide →
[component or pattern]

$ak:shader

/ak:shader [effect or pattern]

Write GLSL fragment shaders for procedural graphics. Topics: shapes (SDF), patterns, noise (Perlin/simplex/cellular), fBm, colors (HSB/RGB), matrices, gradients, animations. Use for generative art, textures, visual effects, WebGL, Three.js shaders.

Open visual guide →
[effect or pattern]

$ak:show-off

/ak:show-off [markdown-or-prompt] [--no-antv|--no-diagram-design|--no-editorial-visuals]

Create preference-aware self-contained HTML pages to showcase work. Use for demos, visual presentations, interactive showcases.

Open visual guide →
[markdown-or-prompt]
--no-antv--no-diagram-design--no-editorial-visuals

$ak:stitch

/ak:stitch [design prompt or action]

AI design generation with Google Stitch. Generate UI designs from text prompts, export Tailwind/HTML/DESIGN.md, orchestrate design-to-code pipeline. Use for rapid prototyping, UI generation, design exploration.

Open visual guide →
[design prompt or action]

$ak:tech-graph

/ak:tech-graph [diagram-type or system description]

Generate production-quality SVG+PNG technical diagrams — architecture, data flow, flowchart, sequence, agent/memory, or concept maps — across 7 visual styles. Use when user wants "generate diagram", "draw diagram", "visualize", "architecture diagram", "flowchart", or any system/flow they want illustrated. Pairs with /ak:preview --diagram for visual self-review and /ak:mermaidjs-v11 for inline-doc diagrams; this skill is the publish-grade output mode.

Open visual guide →
[diagram-type or system description]

$ak:threejs

/ak:threejs [3D scene or feature]

Build 3D web experiences with Three.js. Use for WebGL/WebGPU scenes, GLTF models, animations, physics, VR/XR. Supports 556 searchable examples.

Open visual guide →
[3D scene or feature]

$ak:ui-styling

/ak:ui-styling [component or layout]

Style UIs with shadcn/ui components (Radix UI + Tailwind CSS). Use for accessible components, themes, dark mode, responsive layouts, design systems, color customization.

Open visual guide →
[component or layout]

$ak:ui-ux-pro-max

/ak:ui-ux-pro-max [component|page|design-system] [--style <style>] [--framework <fw>]

UI/UX design intelligence for web and mobile: style selection, color systems, typography, layout, accessibility, interaction states, responsive behavior, forms, charts, design systems, and code review across React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, and HTML/CSS.

Open visual guide →
[component|page|design-system]
--framework--style

$ak:web-design-guidelines

/ak:web-design-guidelines [file-or-pattern]

Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".

Open visual guide →
[file-or-pattern]

$ak:webmcp

/ak:webmcp [page, form, or feature to expose as a tool]

Build agent-ready websites with WebMCP — expose page features as callable tools to in-browser AI agents via document.modelContext. Use whenever the user wants to make a website or web app agent-actionable, add AI-agent tools to a page, make a form agent-callable, use registerTool/getTools/executeTool, the imperative or declarative WebMCP API, tool annotations, the tools Permissions-Policy, or WebMCP evals. In-browser page tools, NOT stdio/HTTP MCP servers — for an MCP server use ak:mcp-builder, to run MCP tools use ak:use-mcp.

Open visual guide →
[page, form, or feature to expose as a tool]

AI Generation & Multimodal

Image/video generation, vision, and media processing

6

$ak:ai-artist

/ak:ai-artist [concept] [--mode search|creative|wild|all] [--provider auto|google|openrouter] [--skip]

Generate product mockups, marketing assets, brand visuals, and concept art via Nano Banana with 129 curated prompts. Mandatory validation interview refines style/mood/colors (use --skip to bypass). 3 modes: search, creative, wild. Styles: Ukiyo-e, Bento grid, cyberpunk, cinematic, vintage patent.

Open visual guide →
[concept]
--mode--provider--skip

$ak:ai-multimodal

/ak:ai-multimodal [file-path] [prompt]

Analyze and generate image, audio, video, and document content through the pinned Multix CLI and live provider catalogs. Use for vision analysis, transcription, OCR, design extraction, and multimodal generation.

Open visual guide →
[file-path][prompt]

$ak:html-video

/ak:html-video [video brief, source URL, repo, template id, or project id]

Create local MP4 videos from HTML/CSS/JS templates with nexu-io/html-video. Covers source checkout setup, template discovery, studio customization, preview, and render verification.

Open visual guide →
[video brief, source URL, repo, template id, or project id]

$ak:hyperframes

/ak:hyperframes [composition or command]

Wrap HeyGen HyperFrames CLI for HTML-first programmatic video. Use for short vertical/social clips, product-launch loops, and motion graphics whose source is HTML composition (not React/Remotion). Local preview + lint, optional HeyGen cloud render.

Open visual guide →
[composition or command]

$ak:media-processing

/ak:media-processing [input-file] [operation]

Process media with FFmpeg (video/audio), ImageMagick (images), RMBG (AI background removal). Use for encoding, format conversion, filters, thumbnails, batch processing, HLS/DASH streaming.

Open visual guide →
[input-file][operation]

$ak:remotion

/ak:remotion [video or component]

Build video content with Remotion in React. Use for programmatic video creation, animated sequences, data-driven video rendering.

Open visual guide →
[video or component]

Build & Ship

Backend, frameworks, deploy, and ship pipelines

13

$ak:agentize

/ak:agentize [feature-or-module] [--both|--mcp|--cli] [--auto|--ask] [--ultra] [--advice] [--yagni]

Convert a codebase, feature, or module into an AI-agent-friendly CLI and/or MCP server. Covers npm packaging, stdio/SSE/Streamable HTTP surfaces, credential resolution, docs, tests, CI, and a companion Claude skill for users who need an existing capability exposed as a reusable agent tool.

Open visual guide →
[feature-or-module]
--advice--ask--auto--both--cli--mcp--ultra--yagni

$ak:backend-development

/ak:backend-development [framework] [task]

Build backends with Node.js, Python, Go (NestJS, FastAPI, Django). Use for REST/GraphQL/gRPC APIs, auth (OAuth, JWT), databases, microservices, security (OWASP), Docker/K8s.

Open visual guide →
[framework][task]

$ak:better-auth

/ak:better-auth [auth-method or feature]

Add authentication with Better Auth (TypeScript). Use for email/password, OAuth providers (Google, GitHub), 2FA/MFA, passkeys/WebAuthn, sessions, RBAC, rate limiting.

Open visual guide →
[auth-method or feature]

$ak:bootstrap

/ak:bootstrap [requirements] [--full|--auto|--fast|--parallel] [--ultra] [--yagni] [--skip-journal]

Bootstrap new projects with research, tech stack, design, planning, and implementation. Modes: full (default interactive), auto (explicit autonomous), fast (skip research), parallel (multi-agent).

Open visual guide →
[requirements]
--auto--fast--full--parallel--skip-journal--ultra--yagni

$ak:databases

/ak:databases [query or schema task]

Design schemas, write queries for MongoDB and PostgreSQL. Use for database design, SQL/NoSQL queries, aggregation pipelines, indexes, migrations, replication, performance optimization, psql CLI.

Open visual guide →
[query or schema task]

$ak:deploy

/ak:deploy [platform] [environment]

Deploy projects to any platform with auto-detection. Use when user says "deploy", "publish", "ship", "go live", "push to production", "host this app", or mentions any hosting platform (Vercel, Netlify, Cloudflare, Railway, Fly.io, Render, Heroku, TOSE, Github Pages, AWS, GCP, Digital Ocean, Vultr, Coolify, Dokploy). Auto-detects deployment target from config files and docs/deployment.md.

Open visual guide →
[platform][environment]

$ak:devops

/ak:devops [platform] [task]

Deploy to Cloudflare (Workers, R2, D1), Docker, GCP (Cloud Run, GKE), Kubernetes (kubectl, Helm). Use for serverless, containers, CI/CD, GitOps, security audit.

Open visual guide →
[platform][task]
/ak:fix COMMON

$ak:fix

/ak:fix [issue] --auto|--review|--quick|--parallel [--ultra] [--advice] [--skip-journal]

Fix bugs, errors, test failures, and CI/CD issues with intelligent routing. Use for type errors, lint issues, log errors, UI bugs, code problems.

Open visual guide →
[issue]
--advice--auto--parallel--quick--review--skip-journal--ultra

$ak:loop

/ak:loop [Goal/Metric description] or inline config block

Autonomous iterative optimization loop — run N iterations against a mechanical metric, learn from git history, auto-keep/discard changes. Use for improving measurable metrics (coverage, performance, bundle size, etc.) through repeated experimentation.

Open visual guide →
[Goal/Metric description]

$ak:mobile-development

/ak:mobile-development [platform] [feature]

Build mobile apps with React Native, Flutter, Swift/SwiftUI, Kotlin/Jetpack Compose. Use for iOS/Android, mobile UX, performance optimization, offline-first, app store deployment.

Open visual guide →
[platform][feature]
/ak:ship COMMON

$ak:ship

/ak:ship [official|stable|main|beta|dev|next] [--both] [--advice] [--merge] [--skip-tests] [--skip-review] [--skip-journal] [--skip-docs] [--social] [--yes-post] [--yes-post-private] [--dry-run]

Ship a completed branch through tests, review, commit, push, and PR creation. Supports official/beta aliases, Kongming advice, optional reviewed merge with CI convergence, dual-target --both, and social posting flags.

Open visual guide →
officialstablemainbetadevnext
--advice--both--dry-run--merge--skip-docs--skip-journal--skip-review--skip-tests--social--yes-post--yes-post-private

$ak:tanstack

/ak:tanstack [framework] [feature]

Build with TanStack Start (full-stack React framework), TanStack Form (headless form management), and TanStack AI (AI streaming/chat). Use when creating TanStack projects, routes, server functions, forms, validation, or AI chat features.

Open visual guide →
[framework][feature]

$ak:web-frameworks

/ak:web-frameworks [framework] [feature]

Build with Next.js (App Router, RSC, SSR, ISR), Turborepo monorepos. Use for React apps, server rendering, build optimization, caching strategies, shared dependencies.

Open visual guide →
[framework][feature]

Git & Version Control

Git ops, GitHub via gh, worktrees, and PR review

4
/ak:git COMMON

$ak:git

/ak:git cm|cp|pr|merge|merge-pr|stack [args]

Git operations with conventional commits. Use for staging, committing, pushing, PRs, merges, and stacked PRs. Auto-splits commits by type/scope. Security scans for secrets.

Open visual guide →
cmcpprmergemerge-prstack
[args]

$ak:github

/ak:github <task description or issue/PR ref> [--interactive] [--advice]

Operate and manage GitHub projects fluently with the gh CLI — create/update/close issues with evidence-backed dedup checks, manage labels, PRs (create, review, rebase, auto-merge), GitHub Projects, Actions CI/CD, and org/repo/environment/secret administration. Use whenever the user asks to file an issue, triage issues, manage a PR lifecycle, inspect CI runs, or administer repositories via gh. For local commits and splits, use /ak:git.

Open visual guide →
<task description or issue/PR ref>
--advice--interactive

$ak:review-pr

/ak:review-pr <PR number or URL> [<PR number or URL> ...] [--fix] [--reply] [--merge] [--advice] [--ultra]

Review one or more GitHub pull requests thoroughly — analyze diffs for correctness, security, breaking changes, code quality, and AI-slop patterns. Accepts multiple PR refs in one call. Auto-detects environments where the GitHub GraphQL API is blocked and falls back to REST. Supports --fix to auto-remediate findings, --reply to post the review, and --merge to merge a ready PR and watch post-merge CI.

Open visual guide →
<PR number or URL>
--advice--fix--merge--reply--ultra

$ak:worktree

/ak:worktree [feature-description] OR [project] [feature]

Create, inspect, and clean isolated git worktrees. Use for feature isolation, worktree health audits, stale cleanup, and monorepo or submodule workflows.

Open visual guide →
createremoveinfoliststatusprune
[feature-description][project][feature]

Test & Debug

Testing, debugging, review, and scenarios

6

$ak:code-review

/ak:code-review [#PR | COMMIT | --pending | codebase [parallel]] [--ultra] [--advice] [--yagni]

Review code quality with evidence-based rigor. Supports input modes: pending changes, PR number, commit hash, and codebase scan. Focuses on bugs, regressions, maintainability, reliability, and verification gaps.

Open visual guide →
[#PR][COMMIT][codebase]
--advice--pending--ultra--yagni

$ak:debug

/ak:debug [error or issue description] [--ultra]

Debug systematically with root cause analysis before fixes. Use for bugs, test failures, unexpected behavior, performance issues, call stack tracing, multi-layer validation, log analysis, CI/CD failures, database diagnostics, system investigation.

Open visual guide →
[error or issue description]
--ultra

$ak:deep-swe

/ak:deep-swe <OpenRouter model slug>

Benchmark a coding model on DeepSWE through Pier and OpenRouter. Use when users ask to run DeepSWE, score a model, or verify coding-agent benchmark results.

Open visual guide →
<OpenRouter model slug>

$ak:scenario

/ak:scenario <file path or feature description> [--iterations N] [--saturation]

Generate comprehensive edge cases and test scenarios by decomposing features across 12 dimensions. Use for pre-implementation risk discovery, QA planning, regression design, and iterative saturation when coverage must be exhaustive.

Open visual guide →
<file path or feature description>
--iterations--saturation
/ak:test COMMON

$ak:test

/ak:test [context] OR ui [url] OR create|optimize|audit [scope] [--advice] [--ultra] [--interview]

Run unit, integration, e2e, and UI tests. Use for test execution, coverage analysis, build verification, visual regression, and QA reports.

Open visual guide →
uicreateoptimizeaudit
[context][url]
--advice--interview--ultra

$ak:web-testing

/ak:web-testing [test-type] [target]

Web testing with Playwright, Vitest, k6. E2E/unit/integration/load/security/visual/a11y testing. Use for test automation, flakiness, Core Web Vitals, mobile gestures, cross-browser.

Open visual guide →
[test-type][target]

Security & Intelligence

Security audits, scans, and threat intel

2

$ak:cti-expert

/ak:cti-expert [target] [--yolo] [--case|--sweep|--query|--flow] [--format html|md] [--no-antv|--no-diagram-design|--no-editorial-visuals]

Analyze cyber threat intelligence and OSINT cases. Use for exposure reviews, domain recon, breach checks, username/email/phone research, image forensics, blockchain tracing, darknet checks, cloud tenant recon, vulnerability lookup, threat modeling, and structured reports.

Open visual guide →
[target]
--case--flow--format--no-antv--no-diagram-design--no-editorial-visuals--query--sweep--yolo

$ak:security

/ak:security [scope] [--secrets-only] [--deps-only] [--fix] [--red-team] [--iterations N]

STRIDE + OWASP-based security audit with optional red-team persona discovery loop and auto-fix. Scans code for vulnerabilities from multiple attacker perspectives (auth attacker, supply chain, insider, infrastructure), categorizes by severity, and can iteratively fix findings using ak:autoresearch pattern.

Open visual guide →
[scope]
--deps-only--fix--iterations--red-team--secrets-only

Docs & Content

Documentation, writing, and content tools

9

$ak:copywriting

/ak:copywriting [copy-type] [context]

Conversion copywriting formulas, headline templates, email copy patterns, landing page structures, CTA optimization, and writing style extraction. Activate for writing high-converting copy, crafting headlines, email campaigns, landing pages, or applying custom writing styles from assets/writing-styles/ directory.

Open visual guide →
[copy-type][context]

$ak:docs

/ak:docs init|update|summarize|agent-context|agents|llms

Analyze a codebase and create, refresh, summarize, or audit project documentation without imposing a fixed docs layout, including authoring and optimizing the root CLAUDE.md/AGENTS.md agent context file, or distilling DO/DON'T rules for that file from git history, CI runs, and (with --source) current source-tree markers.

Open visual guide →
initupdatesummarizeagent-contextagentsllms

$ak:docs-seeker

/ak:docs-seeker [library-name] [topic]

Search library/framework documentation via llms.txt (context7.com). Use for API docs, GitHub repository analysis, technical documentation lookup, latest library features.

Open visual guide →
[library-name][topic]

$ak:document-skills

/ak:document-skills <docx|pdf|pptx|xlsx> [path] [action]

Read, create, and edit Office documents (docx, pdf, pptx, xlsx). Use for tables, forms, slides, spreadsheets.

Open visual guide →
<docx|pdf|pptx|xlsx>[path][action]

$ak:folder-context

/ak:folder-context <target-folder>

Create a compact subfolder CLAUDE.md and linked AGENTS.md for durable local conventions. Use when a folder needs agent context beyond the project root.

Open visual guide →
<target-folder>

$ak:interview-docs

/ak:interview-docs <vision | document-path | topic>

Extract a user's vision and decisions into durable project documents through a guided interview. Use for README, ADR, strategy, principles, and structured-doc authoring.

Open visual guide →
<vision | document-path | topic>

$ak:llms

/ak:llms [path|url] [--full] [--output path]

Generate llms.txt files from docs or codebase scanning. Follows llmstxt.org spec. Use for LLM-friendly site indexes, documentation summaries, AI context optimization.

Open visual guide →
[path|url]
--full--output

$ak:markdown-novel-viewer

/ak:markdown-novel-viewer [file-or-directory]

View markdown files in a calm, book-like reader served via HTTP. Use for long-form content review — RFCs, runbooks, design docs, reports, specs, novels — anywhere you want a distraction-free reading mode in the browser.

Open visual guide →
[file-or-directory]

$ak:mintlify

/ak:mintlify [task] [path]

Build and maintain Mintlify documentation sites. Covers docs.json, MDX components, navigation, page frontmatter, theming, OpenAPI/AsyncAPI, AI docs assets such as llms.txt and skill.md, deployment targets, and local validation CLI commands.

Open visual guide →
[task][path]

MCP & Integration

MCP servers, agents, payments, and integrations

8

$ak:agent-browser

/ak:agent-browser [url or task]

Browser and desktop automation through the agent-browser CLI. Use for long autonomous browsing, compact page snapshots, screenshots, form filling, login/test flows without real Chrome profile state, scraping, exploratory QA, cloud browsers, Electron apps, Slack automation, and bug hunts.

Open visual guide →
[url or task]

$ak:chrome-profile

/ak:chrome-profile [url] [--profile <name>] [--tab <title|url>]

Target a real Google Chrome profile for browser automation through Chrome DevTools MCP. Provides the chrome-profile CLI, profile discovery, live DevTools probing guidance, setup playbooks, and URL-anchor tab selection.

Open visual guide →
[url]
--profile--tab

$ak:google-adk-python

/ak:google-adk-python [agent or feature]

Build AI agents with Google ADK Python. Multi-agent systems, A2A protocol, MCP tools, workflow agents, state/memory, callbacks/plugins, Vertex AI deployment, evaluation.

Open visual guide →
[agent or feature]

$ak:mcp-builder

/ak:mcp-builder [service or API to integrate]

Build MCP servers for LLM-external service integration. Use for FastMCP (Python), MCP SDK (Node/TypeScript), tool design, API integration, resource providers.

Open visual guide →
[service or API to integrate]

$ak:payment-integration

/ak:payment-integration [provider] [task]

Integrate payments with SePay (VietQR), Polar, and Stripe. Checkout, webhooks, subscriptions, QR codes, and multi-provider orders.

Open visual guide →
[provider][task]

$ak:shopify

/ak:shopify [extension-type] [feature]

Build Shopify apps, extensions, themes with Shopify CLI. Use for GraphQL/REST APIs, Polaris UI, Liquid templates, checkout customization, webhooks, billing integration.

Open visual guide →
[extension-type][feature]

$ak:team

/ak:team <template> <context> [--devs|--researchers|--reviewers N] [--delegate]

Orchestrate Agent Teams for parallel multi-session collaboration. Use for research, implementation, review, and debug workflows requiring independent teammates.

Open visual guide →
<template><context>
--delegate--devs--researchers--reviewers

$ak:use-mcp

/ak:use-mcp [task]

Discover and execute MCP server tools through runtime-native capabilities or deterministic direct scripts. Use for MCP integrations, controlled tool execution, capability discovery, and persistent tool catalogs.

Open visual guide →
[task]

Session Management

Handoffs, handover, orchestration, journals, and context

7

$ak:context-engineering

/ak:context-engineering [topic or question]

Check context usage limits, monitor time remaining, optimize token consumption, debug context failures. Use when asking about context percentage, rate limits, usage warnings, context optimization, agent architectures, memory systems.

Open visual guide →
[topic or question]

$ak:handoff

/ak:handoff [task focus] [--output PATH] [--include-diff] [--include-status] [--force]

Create a concise, redacted conversation handoff for a fresh agent session. Use when switching context, ending a work session, or preserving decisions and blockers.

Open visual guide →
[next-session focus]
--force--include-diff--include-status--output

$ak:handover

/ak:handover [task] --agent <id> [--cwd PATH] [--task TEXT] [--handoff PATH] [--model NAME] [--yes]

Hand off in-progress work to a specifically selected coding agent by first capturing a portable handoff, then dispatching one single-job orchestration spec that points that agent at the captured artifact. Thin composition over /ak:handoff and /ak:orchestrate. Not for multi-job orchestration (use /ak:orchestrate) and not capture-only (use /ak:handoff).

Open visual guide →
[task]--agent <id>
--agent--cwd--handoff--model--task--yes

$ak:journal

/ak:journal [topic or reflection]

Write chronological technical journals for session reflection and change analysis. Journals preserve work history; they do not replace current docs or ADRs.

Open visual guide →
[topic or reflection]

$ak:orchestrate

/ak:orchestrate <job-spec.yaml | task description | --resume <run-dir>> [--yes] [--internal]

Coordinate staged or parallel jobs across live-verified coding-agent runtimes and in-session subagents, using capability- and risk-based routing, worktree-isolated writes, resumable state, capture, safety gates, and independent arbiter review.

Open visual guide →
<job-spec.yaml | task description | --resume <run-dir>
--internal--resume--yes

$ak:sumup

/ak:sumup [task|PR|branch|recent]

Summarize completed implementation, failures, workarounds, decisions, behavior, architecture, usage, follow-ups, and next steps. Use after implementation or for a technical recap.

Open visual guide →
[task|PR|branch|recent]

$ak:watzup

/ak:watzup [branch|worktree|status]

Generate short handoff reports from Git branches, remote refs, worktrees, unfinished plans, and roadmap docs. Surfaces priority-ranked next steps with checkbox progress and rationale. Use when the user asks what's in flight, wants progress/next steps, is in a fresh worktree or detached checkout, or needs end-of-session status.

Open visual guide →
[branch|worktree|status]

Project Tracking

Progress, kanban, and project organization

4

$ak:plans-kanban

/ak:plans-kanban [deprecated flags are accepted with warnings]

Open the AgentKit plans dashboard in the CLI config UI. Use for plan kanban views, progress tracking, timeline checks, and quick navigation into plan files.

Open visual guide →
[deprecated flags are accepted with warnings]

$ak:project-management

/ak:project-management [task: status, hydrate, sync, report]

Track progress, update plan statuses, coordinate runtime work, generate reports, and preserve cross-session continuity.

Open visual guide →
[task: status, hydrate, sync, report]

$ak:project-organization

/ak:project-organization [directories or files to organize]

Organize files, directories, and content structure in any project. Use when creating files, determining output paths, organizing existing assets, or standardizing project layout.

Open visual guide →
[directories or files to organize]

$ak:retro

/ak:retro [timeframe] [--compare] [--team] [--format html|md] [--no-antv|--no-diagram-design|--no-editorial-visuals]

Generate data-driven sprint retrospectives from any git history. Use for sprint reviews, commit analysis, code-health indicators, team-velocity reporting, and quarterly engineering reviews. Works on solo or team repos.

Open visual guide →
[timeframe]
--compare--format--no-antv--no-diagram-design--no-editorial-visuals--team

Utilities

ak CLI, skill tools, packaging, and general utilities

11

$ak:agentkit

/ak:agentkit [task to route]

Task router for AgentKit installs. Classifies the task, activates the right installed skills, chains them into the shortest workflow that fits, and spawns installed subagents at defined trigger points to raise output quality. Use at the start of multi-step, multi-domain, or ambiguous work, or when unsure which skill or agent applies.

Open visual guide →
[task to route]

$ak:ak

/ak:ak [goal or subcommand]

Operate the ak control-plane CLI itself — the AgentKit binary that installs, inspects, updates, recovers, and removes kits and their skills. Use when the next action is invoking an ak subcommand (init, kit, skills, plan, journal, doctor, recover, self-update, login), deciding between read-only inspection and lifecycle mutation, disambiguating project vs global scope, or interpreting ak --json output. Not for authoring skills (use /ak:skill-creator) or routing generic work (use /ak:agentkit).

Open visual guide →
[goal or subcommand]

$ak:common

/ak:common [module|helper]

Shared utilities and conventions used across other skills. Internal — invoke only when another skill loads it.

Open visual guide →
[module|helper]

$ak:explain

/ak:explain [subject|path|URL] [--html] [--eli5]

Explain concepts, code, systems, architecture, errors, documents, and technical decisions accurately and at the requested depth. Use whenever the user asks "explain this", "how does this work", "why does X happen", requests a code walkthrough, or needs a complex topic simplified; use --eli5 for plain analogies without losing safety warnings and --html for a visual self-contained explanation.

Open visual guide →
[subject|path|URL]
--eli5--html

$ak:find-skills

/ak:find-skills [capability or task description]

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.

Open visual guide →
[capability or task description]

$ak:gkg

/ak:gkg [symbol or query]

Semantic code analysis with GitLab Knowledge Graph. Use for go-to-definition, find-usages, impact analysis, architecture visualization. Supports Ruby, Java, Kotlin, Python, TypeScript/JavaScript.

Open visual guide →
[symbol or query]

$ak:graphify

/ak:graphify [path] [--mcp|--report|--watch]

Build queryable knowledge graphs from code, docs, papers, and images. Use for codebase understanding, architecture analysis, cross-file relationship discovery, token-efficient navigation.

Open visual guide →
[path]
--mcp--report--watch

$ak:help

/ak:help [topic|skill|command]

Open the AgentKit help index. Use when users ask how to use ak, what skills are available, or which workflow to run.

Open visual guide →
[topic|skill|command]

$ak:repomix

/ak:repomix [path] [--style xml|markdown|plain|json]

Pack repositories into AI-friendly files with Repomix (XML, Markdown, plain text). Use for new-project onboarding, codebase snapshots, LLM context preparation, security audits, third-party library analysis.

Open visual guide →
[path]
--style

$ak:skill-creator

/ak:skill-creator <create|update|audit|optimize> [skill-name|path|kit|--all] [--kit <kit>|--project|--user] [--long-horizon] [--apply] [--from-audit <report>] [--advice]

Create or update Claude skills. Use for new skills, skill scripts, references, packaging, metadata validation, and extending Claude's capabilities.

Open visual guide →
[skill-name|path|kit|--all]
--advice--all--apply--from-audit--kit--long-horizon--project--user

$ak:xia

/ak:xia <github-url-or-owner/repo|local-path> [feature] [--compare|--copy|--improve|--port] [--auto|--fast]

Extract, compare, port, or adapt a feature from a GitHub repository or local repo path into the current project. Use when the user wants to copy behavior from another repo, study how another codebase implements something, compare implementations, or rewrite a feature in the local stack. Triggers on: 'port from', 'copy from repo', 'like how X does it', 'clone feature from', 'adapt from', 'bring feature from', 'borrow from', 'take from repo', 'xia', 'xi a', 'xia feature'.

Open visual guide →
local-path
[feature]
--auto--compare--copy--fast--improve--port

Related guides