About stylesafe
stylesafe is an MCP server and CLI for AI coding agents and CI workflows. It detects CSS cascade and Tailwind utility conflicts before changes ship, returning structured risk reports.
Key features
- 01Detect duplicate declarations and dead rules
- 02Find accidental style overrides
- 03Identify conflicting Tailwind utilities
- 04Understand cn(), clsx(), and twMerge()
- 05Produce structured risk reports
Use cases
- 01Check styles generated by coding agents
- 02Gate conflicting changes in PRs or CI
- 03Audit existing project styles
MCP support
As an MCP server, it provides CSS and Tailwind conflict checks to MCP clients, returning structured reports with confidence, risk scores, and next steps.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEStyleSafe - MCP and CLI
An MCP server and CLI that catches CSS cascade conflicts and Tailwind utility clashes before they ship — built for AI coding agents and CI workflows.
What it does
- Detects duplicate CSS declarations and dead rules from specificity conflicts
- Flags silent style overrides that may be accidental
- Finds conflicting Tailwind utilities (padding, margin, background, shadow, opacity, border, ring, etc.)
- Understands
cn(),clsx(), andtwMerge()call expressions, not just plainclassNamestrings - Returns structured agent-friendly reports with
confidence,riskScore, andnextStep
Quick start
As a CLI tool
npm install -g stylesafe
stylesafe src/components/Button.jsx
stylesafe --changed --fail-on-issues
stylesafe --projectRoot src
stylesafe --watch src
As an MCP server
{
"mcpServers": {
"stylesafe": {
"command": "node",
"args": ["/absolute/path/to/server.js"]
}
}
}
Example output
stylesafe examples/tailwind-conflict.jsx
Returns a structured report:
{
"totalIssues": 5,
"riskScore": 275,
"averageRiskScore": 55,
"riskLevel": "medium",
"passed": true,
"clean": false,
"files": [
{
"filename": "examples/tailwind-conflict.jsx",
"issues": [
{
"type": "tailwind-conflict",
"category": "padding",
"classes": ["p-4", "px-8"],
"confidence": "medium",
"riskScore": 55,
"nextStep": "Choose one utility from the conflicting set or split the classes by scope.",
"requiresUserConfirmation": true
}
]
}
]
}
passed: true means no hard errors. clean: true means zero issues of any kind.
Use cases
- AI code generation: Catch style bugs before the agent commits changes
- PR gate: Block merged changes that introduce cascade conflicts or utility clashes
- Agent workflows: Integrate as a confirmation step in multi-turn coding tasks
- Style audits: Run across a project to find existing issues
CLI options
stylesafe <file>— analyze a single filestylesafe <dir>— analyze a directory recursivelystylesafe --projectRoot <dir>— same as abovestylesafe --changed— analyze git-changed files (for PR/CI)stylesafe --changed --fail-on-issues— exit with code 1 if any issues foundstylesafe --watch src— re-run on every file changenpm test— run regression tests
Config
Place a .styleintegrityrc file in your project root:
{
"ignore": ["dist", "*.min.css"],
"failOn": ["error", "warning"]
}
GitHub Actions
Add to .github/workflows/style-check.yml:
name: stylesafe
on:
pull_request:
push:
branches: [main]
jobs:
stylesafe:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: actions/setup-node@v4
with:
node-version: 20
- run: npm install
- run: node server.js --changed --fail-on-issues
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.