Skip to content
Get packs

Guide · Open source

How to use Claude-Red: install the offensive security skills for Claude

Updated · written with AI from the project's own documentation and checked against our catalogue data

claude-red is an open-source collection of SKILL.md files for the Claude skills system. Each file gives Claude structured methodology for one offensive security topic, such as SQL injection or Active Directory attacks. This guide explains how to install it, how skills load, and how to use it for authorized work.

claude-redSnailSploit/Claude-Red · ★ 10.2k · MIT

Quick start

  1. 1Confirm your Claude access. Make sure you can use Claude through the Claude skills system, Claude Code, or Claude.ai.
  2. 2Clone the repository. Run the git clone command to place claude-red in ~/.claude/skills/claude-red.
  3. 3Pick a route. Use the skills directory for automatic loading, or paste a SKILL.md into a Claude.ai Project.
  4. 4Optionally install one category. Use the sparse checkout or the install script with --category to fetch only what you need.
  5. 5Ask about a specific topic. Mention the topic, such as SQL injection, so Claude loads the matching skill.
  6. 6Work within authorization. Apply the methodology only to systems you are authorized to test.

What claude-red is and who it is for

claude-red is a curated library of offensive security skills designed for the Claude skills system. Each skill is a structured SKILL.md file that primes Claude with methodology for a specific attack surface, covering topics from SQL injection to shellcode and EDR evasion to exploit development.

The README describes its intended use as authorized red team engagements, bug bounty triage, security research, CTF preparation, operator training and methodical attack surface exploration. It is aimed at people who already use Claude.

  • Security professionals who want topic-specific methodology for authorized testing
  • Students preparing for CTFs or learning offensive security concepts
  • Bug bounty hunters who want structured triage checklists for web vulnerability classes
  • Red team operators who want a reference for specific attack surfaces

What you need before you install

You need access to Claude through one of the three routes the README names: the Claude skills system, Claude Code, or Claude.ai. The files are built for the Claude skills system, so people who do not use Claude will not get value from them.

The README does not describe bundled scanners or binaries. claude-red is a set of prompt-style guidance files, not a tool you run against a target.

Install claude-red into the Claude skills directory

The recommended route is to clone the repository into your Claude skills directory. This is the clone-and-build method from the README.

Claude auto-loads matching skills based on conversational triggers. For example, mentioning SQL injection loads the offensive-sqli skill.

git clone https://github.com/SnailSploit/claude-red ~/.claude/skills/claude-red

Install only one category

If you do not want the whole library, the README shows a sparse checkout that fetches only selected folders. The example below pulls the web and Active Directory categories.

git clone --filter=blob:none --sparse https://github.com/SnailSploit/claude-red cd claude-red && git sparse-checkout set Skills/web Skills/active-directory

Use claude-red in Claude Code, Claude.ai or the install script

In Claude Code, the README pipes a SKILL.md file into claude with the system-file option. In Claude.ai, paste the contents of a SKILL.md into a Project's system prompt or prepend it to your conversation.

The repository also includes an install script. Its options are shown below: interactive mode, an explicit target directory, or a single category.

  • Interactive mode: ./install.sh
  • Explicit target: ./install.sh --target ~/.claude/skills
  • Single category: ./install.sh --category web
cat Skills/web/offensive-sqli/SKILL.md | claude --system-file -

Find the right skill for your task

The repository groups its skills into categories such as Web Application, Auth and Identity, Active Directory, Wireless, Cloud and Exploit Development. Each category has a skill index with a short description of each file.

For example, the Web Application category includes offensive-sqli, offensive-xss and offensive-ssrf. The Active Directory category has a single offensive-active-directory skill covering Kerberoast, ASREProast, ACL abuse and ADCS.

  • Browse the categories table in the README to find your attack surface
  • Check the skill index for a one-line description of each file
  • Use the trigger word for a topic to load its skill in Claude

What people use it for

  • Structured triage checklists for web vulnerability classes during a bug bounty program
  • Methodology review for an authorized Active Directory assessment
  • Preparing for a CTF by working through a specific attack surface with Claude
  • Building a reference for wireless testing topics such as WPA2-PSK and WPA3-SAE

Frequently asked questions

Is claude-red free to use?

claude-red is released under the MIT licence, as shown in the README. Check the LICENSE file in the repository for the full terms.

What do I need to install claude-red?

You need access to Claude through the Claude skills system, Claude Code, or Claude.ai. You also clone the repository with git, as the README shows.

Does claude-red include hacking tools?

The README does not describe bundled scanners or binaries. It describes a collection of SKILL.md methodology files for Claude.

How do I install claude-red on Windows or Mac?

The README gives the clone command and the install script but does not describe platform-specific steps. Use the git command shown in the guide and check the repository for any platform notes.

Can I use claude-red without the Claude skills system?

Yes. The README describes pasting a SKILL.md into a Claude.ai Project's system prompt or prepending it to a conversation, or piping it into Claude Code.

Is claude-red suitable for beginners?

It is prompt-style guidance rather than hands-on tools. The README frames it around authorized engagements, so beginners should understand the methodology and the authorization requirement before using it.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

claude-red suits security professionals, students and bug bounty hunters who already use Claude and want topic-specific methodology for authorized testing. If you need hands-on scanners or do not use Claude, pick a different kind of tool.

Install options & downloads