v1.7.0 — 5 new agents & requirements skills

Make AI assistants
actually useful.

One command. AI Kit scans your project, detects the stack, and generates tailored rules, skills, and agents for Claude Code, Cursor, and 3 more tools. Zero config. Project-aware from the first conversation.

$npx @mikulgohil/ai-kit init
Get StartedRead the DocsGitHub
48
Pre-built Skills
16
Specialized Agents
5+
AI Tools Supported
0 min
Setup Time
Features

Everything your AI assistant needs to know

AI Kit generates persistent, project-aware configuration so every conversation starts informed.

⚡

Auto Stack Detection

Scans your package.json, config files, and directory structure. Detects framework, router, CMS, styling, TypeScript config, monorepo setup, and package manager — then generates rules specific to your exact stack.

✓Next.js 15 — App Router detected
✓TypeScript — strict mode enabled
✓Tailwind CSS v4 — @theme tokens
✓Turborepo — 3 workspaces found
✓Sitecore XM Cloud — Content SDK v2
✗Figma — not detected
⌘

48 Pre-Built Skills

Structured AI workflows — /review, /new-component,/security-check, /deep-interview and 44 more. No prompt writing needed.

✦

16 Specialized Agents

Delegate to focused AI agents — planner, code-reviewer, security-reviewer, build-resolver, data-scientist, api-designer, and more.

↻

Automated Quality Hooks

Auto-format, type-check, lint warnings, and git safety — running in real-time as the AI writes code. Three profiles: minimal, standard, strict.

🛡

Built-in Security Audit

Scans for exposed secrets, OWASP risks, XSS vectors, and misconfigurations. A-F health grade with actionable recommendations.

⭐

5+ AI Tools from One Source

Generates configs for Claude Code, Cursor, Windsurf, Aider, and Cline from a single scan. Switch tools without losing context.

📈

AI That Improves Over Time

Mistakes log auto-captures build failures. Decisions log tracks technical choices. The AI references both — getting smarter every session.

🔍

Component Scanner

Discovers all React components and generates .ai.md docs with health scores, props tables, and Sitecore field mappings.

💰

Token Tracking & Cost Estimates

Daily/weekly/monthly usage summaries, per-project cost breakdown, budget alerts, model recommendations, and ROI estimates.

How It Works

30 seconds to project-aware AI

No manual configuration. No YAML files. No prompt engineering.

1

Run init

npx @mikulgohil/ai-kit init in any project directory

2

Auto-detect

AI Kit scans your stack, configs, and project structure

3

Generate

Tailored rules, skills, agents, hooks, and guides appear

4

Commit & go

Commit the files — every developer gets project-aware AI instantly

Multi-Tool

One scan. Every AI tool.

Generate configs for all major AI coding assistants from a single source of truth.

💬
Claude Code
CLAUDE.md + skills + agents
◎
Cursor
.cursorrules + .mdc files
🌊
Windsurf
.windsurfrules
🤖
Aider
.aider.conf.yml
⚙
Cline
.clinerules
Agents

16 specialists. Always on call.

Delegate complex tasks to focused AI agents with domain expertise.

📋

@planner

Break features into implementation plans with dependencies and risk assessment

🔎

@code-reviewer

Deep quality review — patterns, performance, types, and conventions

🛡

@security-reviewer

OWASP Top 10, XSS, CSRF, secrets detection, and auth flow analysis

🔧

@build-resolver

Diagnose and fix build errors, type conflicts, and dependency issues

📝

@doc-updater

Keep documentation in sync with code changes automatically

✂

@refactor-cleaner

Find and remove dead code, unused imports, and unnecessary complexity

🧪

@tdd-guide

Test-driven development workflow — red, green, refactor with guidance

🚨

@ci-debugger

Analyze CI/CD failures, parse logs, and suggest targeted fixes

🎭

@e2e-runner

Playwright tests with Page Object Model and smart selectors

🏛

@sitecore-specialist

XM Cloud patterns, Content SDK v2, Experience Edge, and field helpers

🏗

@architect

SSR/SSG/ISR strategy, component hierarchy, data flow, and rendering patterns

📊

@data-scientist

ML pipelines, model evaluation, data analysis, and experiment tracking

⚡

@performance-profiler

Core Web Vitals, bundle analysis, runtime profiling, and rendering optimization

🔄

@migration-specialist

Framework upgrades, breaking change detection, codemods, and incremental adoption

🔍

@dependency-auditor

Vulnerability scanning, outdated packages, license compliance, and bundle impact

🌐

@api-designer

REST/GraphQL API design, schema validation, versioning, and error handling

Before & After

The impact is measurable

Real numbers from teams using AI Kit in production.

MetricWith AI KitWithout AI Kit
Context setup per conversation0 min — auto-loaded5-10 min re-explaining
Code review cycles per PR1-2 rounds2-4 rounds
Component creation time10-15 min30-60 min
New developer onboarding2-3 days1-2 weeks
Security issues caughtAt development timeAt PR review or production
Knowledge retentionLogged in decisions & mistakesLost when developers leave
AI tool switching costZero — same rules, 5+ toolsStart over from scratch
CLI

13 commands. Full control.

From initial setup to ongoing maintenance — everything from the terminal.

▶

ai-kit init

Scan project and generate all configs

↻

ai-kit update

Re-scan and refresh (preserves custom edits)

❤

ai-kit health

One-glance A-F project health dashboard

🛡

ai-kit audit

Security and configuration health check

🩺

ai-kit doctor

Diagnose setup issues and misconfigurations

📊

ai-kit stats

Project complexity metrics and analysis

💰

ai-kit tokens

Token usage summary and cost estimates

🔍

ai-kit diff

Preview what would change on update

📦

ai-kit export

Export rules to Windsurf, Aider, Cline

🎨

ai-kit patterns

Generate pattern library from codebase

✂

ai-kit dead-code

Find unused components and dead code

📏

ai-kit drift

Detect code vs documentation drift

🗑

ai-kit reset

Remove all AI Kit generated files cleanly

FAQ

Common questions

Quick answers. Full FAQ in the docs.

Yes. AI Kit is 100% free and open source under the MIT license. It works with any AI coding tool — free or paid.
Yes — that's actually the sweet spot. The $20 plan gives you enough tokens to be productive but not enough to waste. AI Kit eliminates the #1 token sink: re-explaining your project every session. Every token goes to actual coding, not context setup.
Yes. AI Kit generates configs for both Claude Code and Cursor (plus Windsurf, Aider, and Cline). Skills work in both tools via auto-discovery — the AI reads skill descriptions and applies the right one automatically.
No. AI Kit only creates configuration files (CLAUDE.md, .cursorrules, skills, agents, etc.). It never touches your source code, dependencies, or existing configs.
About 30 seconds. Run npx @mikulgohil/ai-kit init, answer a few prompts (tool choice, strictness level), and everything is generated. Zero manual configuration.
No. The generated files are static markdown read once at session start. No runtime cost, no background processes, no API calls. It's just context — the AI loads it and forgets the file exists.
Yes. Add your own rules outside the AI-KIT markers — they're preserved when you run ai-kit update. You can also add custom skills and agents that AI Kit won't overwrite.
AI Kit still generates base rules (coding standards, naming, testing, security, docs) that apply to any project. Framework-specific fragments are additive — they're only included when detected.

Ready to make AI actually useful?

One command. 30 seconds. Your AI assistant goes from generic to project-aware.

$npx @mikulgohil/ai-kit init