SnailSploit/Claude-Red — 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 expert-level methodology for a specific attack surface — from SQLi to shellcode, EDR evasion to exploit development.

What happened

A developer shared claude-red on GitHub. It is a library of 78 'skills' for the Claude skills system. Each skill is a SKILL.md file. The file gives Claude expert steps and know-how for one type of security attack, such as SQL injection or EDR evasion. Claude loads a skill when the chat mentions a matching topic.

Key numbers and facts

• 78 skills in total
• 23 categories
• Categories include web apps, wireless, cloud, mobile, IoT, and exploit development
• Web Application has 16 skills; Wireless has 14
• Licensed under MIT

Background for newcomers

Offensive security means testing computer systems by attacking them, so the owner can find and fix weak spots. Teams that do this work are called red teams. The project says it is meant for authorized tests, bug bounty work (paid programs that reward people for finding bugs), security research, and training. Skills are loaded only when needed, so they do not use up extra chat memory.

Why it matters

This shows how AI assistants can be given deep expert knowledge in a narrow field with simple text files. It also raises a question: the same help that aids a legitimate tester could also help a bad actor. The project's stated purpose is authorized testing, but the tool itself does not check who is using it.

What to watch

• How many people install and use the skills
• Whether the project adds more skills, as its roadmap suggests
• How AI companies respond to security tools built on their skills systems