SnailSploit/Claude-Red

What it claims

claude-red by Kai Aizen (SnailSploit) is a set of SKILL.md files that give Claude methodology for specific attack surfaces, from SQL injection to shellcode and ADCS abuse. The README lists 23 category directories under Skills/: web, auth, active-directory, wireless, cloud, mobile, iot, infrastructure, exploit-dev, fuzzing, recon, api, container, cicd, crypto, privesc, post-exploitation, forensics, supply-chain, social-engineering, network, ai and utility. Many skills build on an offensive checklist collection by Sahar Shlichov. The stated uses are authorized red team engagements, bug bounty triage, security research, CTF preparation and operator training.

How it works

Each skill is named for its surface and loads when the conversation mentions it. For example, mentioning SQL injection loads offensive-sqli, at Skills/web/offensive-sqli/SKILL.md. Contributions follow a skill template and frontmatter standard in CONTRIBUTING.md. The project prefers focused, single-surface skills over broad overviews.

How to use it

Full install into the Claude Skills directory, the recommended path:

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

Sparse checkout of chosen categories, then feeding skills to the CLI as a system file:

git clone --filter=blob:none --sparse https://github.com/SnailSploit/claude-red
cd claude-red && git sparse-checkout set Skills/web Skills/active-directory
cat Skills/web/offensive-sqli/SKILL.md | claude --system-file -
cat Skills/active-directory/**/SKILL.md | claude --system-file -

The installer script:

./install.sh                           # interactive
./install.sh --target ~/.claude/skills # explicit target
./install.sh --category web            # single category

Without the Skills system, a SKILL.md can be pasted into a Project's system prompt or prepended to a conversation. Trigger-based loading does not apply in that case, so the pasted text uses context the whole time.

Limits

The library is unfinished. About 130 skills is a target spread over several release phases, and the README does not say how many skills exist now. CHANGELOG.md has the release history. The page shows no skill contents beyond one file path, and it gives no examples or evaluation of how the skills change Claude's output. The licence is MIT, with attribution requested but not required.