styleguard

Your linter checks syntax. StyleGuard checks consistency.

Website · Quick Start · Patterns · Pricing

---

ESLint checks if your code is valid. StyleGuard checks if it is consistent.

A file where half the variables are camelCase and the other half are snake_case. A function with 8 parameters and 6 levels of nesting. A magic number 86400 buried in a conditional. Imports scattered in random order. Three different quoting styles in the same module.

Formatters like Prettier fix whitespace. Linters like ESLint catch syntax errors. But neither catches consistency violations -- the naming convention drift, magic number creep, and structural decay that make codebases unreadable over time.

StyleGuard scans your code for 90+ style consistency patterns across 5 languages. Locally. Before you commit.

Quick Start

# Install via ClawHub (free)
clawhub install styleguard

# Scan your codebase
styleguard scan

# Scan a specific directory
styleguard scan src/

# Scan a single file
styleguard scan app/services/auth.ts

# Install pre-commit hooks (Pro)
styleguard hooks install

# Generate a style audit report (Pro)
styleguard report

# Auto-fix suggestions (Pro)
styleguard fix

# Enforce team style policies (Team)
styleguard policy

# Export SARIF for GitHub Code Scanning (Team)
styleguard sarif

# Capture style baseline (Team)
styleguard baseline

What It Catches

StyleGuard runs 90+ checks organized in 6 categories:

Naming Conventions (22 checks)

Check	Description	Severity
Mixed Case Styles	camelCase and snake_case mixed in one file	High
PascalCase Violations	Classes/components not PascalCase	High
UPPER_CASE Constants	Constants not UPPER_SNAKE_CASE	Medium
File Naming Inconsistency	Mixed kebab-case/camelCase/snake_case filenames	Medium
Single-Letter Variables	Variables like x, d, t outside loops	Medium
Misleading Booleans	Booleans without is/has/should/can prefix	Medium
Hungarian Notation	Type prefix remnants (strName, bFlag)	Low
Event Handler Names	Mixed handleClick vs onClick conventions	Medium
Acronym Casing	URL vs Url vs url inconsistency	Medium

Function Quality (18 checks)

Check	Description	Severity
Long Functions	Functions exceeding 50 lines	High
Too Many Parameters	Functions with > 5 params	High
Deep Nesting	> 4 levels of indentation	Critical
God Functions	Functions with too many responsibilities	Critical
Missing Early Returns	Deep if/else chains	Medium
Callback Hell	> 3 levels of nested callbacks	Critical
Arrow vs Declaration	Inconsistent function declaration styles	Low
Multiple Return Types	Functions returning mixed types	Medium

Magic Numbers & Strings (14 checks)

Check	Description	Severity
Magic Numbers	Raw numbers in conditions (except 0, 1, -1)	Medium
Repeated Strings	Same string literal used > 2 times	Medium
Hardcoded URLs	URLs embedded in business logic	High
Hardcoded Timeouts	Magic delay/timeout values	Medium
Magic Array Indices	Accessing arr[7] without explanation	Low
Hardcoded Colors	Hex colors in logic files	Low
Numeric Status Codes	404, 500 without named constants	Medium

Import & Module Organization (16 checks)

Check	Description	Severity
Unsorted Imports	Imports not in alphabetical order	Low
Mixed Import Styles	require() + import in same file	High
Circular Import Hints	Patterns suggesting circular deps	Critical
Wildcard Imports	import * from module	Medium
Deep Relative Paths	../../../ depth > 3 levels	Medium
Unused Imports	Imported names not referenced	High
Side-Effect Position	Side-effect imports not at top	Medium

Comment & Documentation (12 checks)

Check	Description	Severity
TODO/FIXME Buildup	Excessive TODO/FIXME/HACK markers	Medium
Commented-Out Code	Large blocks of dead code in comments	Medium
Dead JSDoc	Docstrings with wrong parameter names	High
ts-ignore Accumulation	Multiple @ts-ignore suppressions	High
eslint-disable Abuse	Suppression without justification	Medium
Empty Catch Blocks	Exceptions silently swallowed	High

Code Consistency (8+ checks)

Check	Description	Severity
Mixed Quotes	Single and double quotes in same file	Medium
Inconsistent Semicolons	Mixed semicolon usage	Medium
Mixed Equality	== and === in same JS/TS file	High
Mixed Async Patterns	Callbacks + promises + async/await	High
Inconsistent Null Checks	Mixed null/undefined/falsy patterns	Medium
Inconsistent Trailing Commas	Mixed trailing comma usage	Low

Supported Languages

Language	Naming	Functions	Magic Numbers	Imports	Comments	Consistency
JavaScript/TypeScript	Yes	Yes	Yes	Yes	Yes	Yes
Python	Yes	Yes	Yes	Yes	Yes	Yes
Go	Yes	Yes	Yes	Yes	Yes	Yes
Java	Yes	Yes	Yes	Yes	Yes	Yes
Ruby	Yes	Yes	Yes	Yes	Yes	Yes

How It Compares

Feature	StyleGuard	ESLint	Prettier	Pylint	RuboCop	gofmt	Checkstyle
Cross-language	5 languages	JS/TS only	JS/TS only	Python only	Ruby only	Go only	Java only
Naming consistency	Yes	Partial	No	Partial	Partial	No	Partial
Magic number detection	Yes	Plugin	No	Yes	Yes	No	Yes
Function quality metrics	Yes	Plugin	No	Yes	Partial	No	Yes
Import ordering	Yes	Plugin	No	Plugin	Partial	Yes	Partial
Cross-file consistency	Yes	No	No	No	No	No	No
Pre-commit hooks	Yes	Plugin	Plugin	Manual	Manual	Manual	Manual
Zero config	Yes	No	Partial	No	No	Yes	No
Offline license	Yes	OSS	OSS	OSS	OSS	OSS	OSS
Setup time	30 seconds	10+ min	5+ min	5+ min	5+ min	Built-in	15+ min
ClawHub integration	Yes	No	No	No	No	No	No
Zero telemetry	Yes	Yes	Yes	Yes	Yes	Yes	Yes

Key distinction: ESLint, Prettier, Pylint, RuboCop, gofmt, and Checkstyle are formatters and single-language linters. StyleGuard catches consistency issues they miss -- naming convention drift across a file, magic numbers in conditionals, function complexity creep, and import disorder -- across all 5 languages with one tool.

Pricing

Feature	Free	Pro ($19/user/mo)	Team ($39/user/mo)
Style scan (90+ checks)	5 files	Unlimited	Unlimited
5 language support	Yes	Yes	Yes
Style score + grade	Yes	Yes	Yes
Detailed remediation advice	Yes	Yes	Yes
Pre-commit hooks		Yes	Yes
Style audit report		Yes	Yes
Auto-fix suggestions		Yes	Yes
Team style policies			Yes
SARIF export			Yes
Style baseline			Yes

Commands

Free Tier (no license required)

Command	Description
styleguard scan [target]	Style scan (up to 5 source files)
styleguard status	Show license and config info
styleguard help	Show help

Pro Tier ($19/user/month)

Command	Description
styleguard scan [target]	Unlimited source file scanning
styleguard hooks install	Install pre-commit style hooks
styleguard hooks uninstall	Remove StyleGuard hooks
styleguard report [dir]	Generate markdown style report
styleguard fix [dir]	Generate auto-fix suggestions

Team Tier ($39/user/month)

Command	Description
styleguard policy [dir]	Enforce team style policies
styleguard sarif [dir]	Export findings in SARIF format
styleguard baseline [dir]	Capture style baseline for incremental adoption

Configuration

Add to ~/.openclaw/openclaw.json:

{
  "skills": {
    "entries": {
      "styleguard": {
        "enabled": true,
        "apiKey": "YOUR_LICENSE_KEY",
        "config": {
          "severityThreshold": "medium",
          "excludePatterns": ["**/node_modules/**", "**/dist/**"],
          "reportFormat": "markdown",
          "maxFunctionLines": 50,
          "maxParameters": 5,
          "maxNestingDepth": 4
        }
      }
    }
  }
}

Or set the environment variable:

export STYLEGUARD_LICENSE_KEY="your-key-here"

Exit Codes

• 0 -- Clean (score >= 70, no critical issues)
• 1 -- Issues found (score < 70 or critical issues present)

Perfect for CI/CD integration.

Privacy

• 100% local -- no code or data sent externally
• Zero telemetry
• Offline license validation (JWT-based)
• Pattern matching only -- no AST parsing or code execution

Part of the ClawHub Ecosystem

StyleGuard is built for ClawHub, the marketplace for developer tools that plug into your CLI. Other skills in the ecosystem:

• PerfGuard -- Performance anti-pattern scanner
• DocSync -- Keep docs in sync with code
• DepGuard -- Dependency audit and vulnerability scanning
• EnvGuard -- Pre-commit secret detection
• APIShield -- API endpoint security auditor
• GitPulse -- Git workflow analytics
• ConfigSafe -- Configuration drift detector
• LicenseGuard -- License compliance scanner

License

MIT