MegaLinter, by

MegaLinter is an open-source tool for CI/CD workflows that analyzes the consistency of your code, IaC, configuration, and scripts in your repository to ensure all your project sources are clean and formatted, no matter which IDE or toolbox is used by your developers. Powered by OX Security.

Supports 67 languages, 23 formats, 22 tooling formats, and is ready to use out of the box as a GitHub Action or with any CI system. It is highly configurable and free for all uses.

MegaLinter has native integrations with popular CI/CD tools and is compliant with most Coding Agents.

• Use the new LLM Advisor that will tell you how to fix linters issues
• Create your own MegaLinter Custom Flavors to improve performance

Before you go further, see the online documentation website, which offers much easier navigation than this README

Why MegaLinter

Projects need to contain clean code in order to avoid technical debt, which makes ongoing maintenance harder and more time-consuming.

By using code formatters and code linters, you ensure that your codebase is easier to read and follows best practices, from kickoff through each step of the project lifecycle.

Not all developers use linters in their IDEs, which makes code reviews harder and longer to process.

By using MegaLinter, you'll enjoy the following benefits for your team:

• At each pull request, it automatically analyzes all updated code across all languages.
• By reading error logs, developers learn best practices for the languages they use.
• The MegaLinter documentation provides a list of IDE plugins for each linter, so developers know which linter and plugins to install.
• MegaLinter works out of the box after a quick setup.
• Formatting and fixes can be automatically applied directly to the Git branch or provided in reports.
• This tool is 100% open source and free for all uses (personal, professional, public, and private repositories).
• MegaLinter can run on any CI tool and be run locally: no need to authorize an external application, and your codebase never leaves your tooling ecosystem.

Join a continuously growing community: MegaLinter Docker images are pulled hundreds of thousands of times every month.

Quick Start

• With a coding agent: install the MegaLinter agent skills with npx skills add oxsecurity/megalinter/skills, then ask your agent to "setup megalinter".
• Or with the interactive wizard: run npx mega-linter-runner --install to generate configuration files (you need Node.js installed).
• Commit, push, and create a pull request.
• Watch!

Notes:

• This repo is a hard fork of GitHub Super-Linter, rewritten in Python to add many additional features.
• If you are a Super-Linter user, you can transparently switch to MegaLinter and keep the same configuration (just replace super-linter/super-linter@v3 with oxsecurity/megalinter@v10 in your GitHub Action YAML file, like on this PR).
• If you want to use MegaLinter's extra features (recommended), please take 5 minutes to use the assisted installation.
• For a beginner-friendly example of getting started with MegaLinter, check out this blog post by Alec Johnson.

Coding Agents

MegaLinter is designed to work hand in hand with your coding agent:

Get started

Install the MegaLinter agent skills at the root of your repository:

With Claude Code (the skills are copied directly into .claude/skills/):

npx skills add oxsecurity/megalinter/skills -s '*' -a claude-code -y

With another coding agent, replace claude-code with your agent's identifier (cursor, github-copilot, codex, antigravity, opencode... full list in the skills CLI documentation) — or let the CLI detect your installed agents and install the skills for all of them at once:

npx skills add oxsecurity/megalinter/skills -s '*' -y --copy

Then just talk to your agent:

• "Setup MegaLinter on this repo"
• "Run MegaLinter and fix the errors"
• "Why is the MegaLinter CI job failing? Fix it"

What the skills do

Sub-agents orchestration

On agents supporting sub-agents (Claude Code, OpenCode, GitHub Copilot custom agents...), megalinter-setup installs three sub-agents that keep large CI logs and linter outputs out of your main agent's context, and run on low-cost models when possible:

flowchart TD
    U(["👤 You: <i>run megalinter and fix the errors</i>"]) --> O["🎯 megalinter<br/>(orchestrator skill)"]
    O -->|"no config found"| S["🛠️ megalinter-setup"]
    S -->|"npx mega-linter-runner --install"| CFG[/".mega-linter.yml<br/>+ CI workflow"/]
    O --> C["🔍 megalinter-check"]
    C -->|"watch mode"| W[["👀 megalinter-watcher<br/>sub-agent, low-cost model"]]
    W --> CI[/"CI job logs<br/>GitHub / GitLab / Azure / Bitbucket"/]
    C -->|"local mode"| RUN[["🐳 megalinter-runner<br/>sub-agent, low-cost model"]]
    RUN --> IMG[/"MegaLinter Docker image<br/>docker or podman"/]
    W --> ERR[/"📋 Compact error list<br/>grouped by linter"/]
    RUN --> ERR
    ERR --> F["🔧 megalinter-fix"]
    F -->|"one sub-agent per failing linter,<br/>in parallel"| FX1[["🤖 megalinter-fixer<br/>+ python_ruff fix guide"]]
    F --> FX2[["🤖 megalinter-fixer<br/>+ markdown_markdownlint fix guide"]]
    FX1 --> RES[/"✏️ Fixed files<br/>+ proposed rule disables"/]
    FX2 --> RES
    RES -->|"you confirm disables"| RC["♻️ Targeted re-check<br/>parallel standalone linter images"]
    RC -->|"errors remain (max 3 loops)"| F
    RC -->|"clean"| DONE(["✅ Commit on a branch<br/>(never on main)"])

    classDef user fill:#e8f5e9,stroke:#2e7d32,stroke-width:2px,color:#1b5e20
    classDef skill fill:#ede7f6,stroke:#5e35b1,stroke-width:2px,color:#311b92
    classDef subagent fill:#e3f2fd,stroke:#1976d2,stroke-width:2px,color:#0d47a1
    classDef artifact fill:#fff8e1,stroke:#f9a825,stroke-width:1px,color:#5d4037
    classDef recheck fill:#e0f2f1,stroke:#00897b,stroke-width:2px,color:#004d40

    class U,DONE user
    class O,S,C,F skill
    class W,RUN,FX1,FX2 subagent
    class CFG,CI,IMG,ERR,RES artifact
    class RC recheck

The fix guides combine information generated from the linter descriptors (auto-fix support, rules documentation URLs, MegaLinter tuning variables) with curated fix, inline-disable and ignore instructions grounded in each linter's official documentation.

Safety rules

• Safe fixes are applied automatically; ambiguous ones are asked to you
• Disabling a linter or a rule always requires your confirmation
• Commits are never pushed to the default branch

See also the installation page for coding agents, or install everything at once as an agent plugin.

Supported Linters

All linters are integrated into the MegaLinter Docker image, which is frequently updated with their latest versions.

Languages

	Language	Linter	Additional
	BASH	bash-exec BASH_EXEC
	BASH	shellcheck BASH_SHELLCHECK
	BASH	shfmt BASH_SHFMT
	C	cppcheck C_CPPCHECK
	C	cpplint C_CPPLINT
	C	clang-format C_CLANG_FORMAT
	CLOJURE	clj-kondo CLOJURE_CLJ_KONDO
	CLOJURE	cljstyle CLOJURE_CLJSTYLE
	COFFEE	coffeelint COFFEE_COFFEELINT	🚫
	C++ (CPP)	cppcheck CPP_CPPCHECK
	C++ (CPP)	cpplint CPP_CPPLINT
	C++ (CPP)	clang-format CPP_CLANG_FORMAT
	C# (CSHARP)	dotnet-format CSHARP_DOTNET_FORMAT
	C# (CSHARP)	csharpier CSHARP_CSHARPIER
	C# (CSHARP)	roslynator CSHARP_ROSLYNATOR
	DART	dartanalyzer DART_DARTANALYZER
	GO	golangci-lint GO_GOLANGCI_LINT
	GO	revive GO_REVIVE
	GROOVY	npm-groovy-lint GROOVY_NPM_GROOVY_LINT
	JAVA	checkstyle JAVA_CHECKSTYLE
	JAVA	pmd JAVA_PMD
	JAVASCRIPT	eslint JAVASCRIPT_ES
	JAVASCRIPT	standard JAVASCRIPT_STANDARD
	JAVASCRIPT	prettier JAVASCRIPT_PRETTIER
	JAVASCRIPT	biome JAVASCRIPT_BIOME
	JSX	eslint JSX_ESLINT

…view the full README on GitHub.