How do I install the Cai skill?

Run npx skillmds@latest add neronain/cai in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Cai?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

neronain (@neronain) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Cai by neronain — Agent Skill

Name: Cai
Author: neronain

Cai

Launch and interact with CAI (Cybersecurity AI) — an AI-powered offensive/defensive security framework with 300+ model support. Use for CTF challenges, penetration testing workflows, reconnaissance, and vulnerability research.

neronain Updated Sep 21, 2026

File contents

CAI — Cybersecurity AI

CAI is an open-source AI-powered security framework by Alias Robotics.

300+ AI models (Anthropic, OpenAI, DeepSeek, Ollama, etc.)
Specialized agents: CTF, recon, red team, boot2root, flag discriminator
Human-in-the-loop via Ctrl+C intervention

Quick Launch

# Verify installation
pip show cai-framework

# Launch interactive shell (binary path on macOS with system Python 3.9)
/Users/tananan/Library/Python/3.9/bin/cai

# Or add to PATH permanently:
# echo 'export PATH="$HOME/Library/Python/3.9/bin:$PATH"' >> ~/.zshrc && source ~/.zshrc
# Then just: cai

# Disable license check if needed
CAI_LICENSE_OFF=1 /Users/tananan/Library/Python/3.9/bin/cai

Agent Selection

Once inside cai, use /agent to list and switch agents:

Agent	Purpose
`ctf_agent`	CTF challenge solving
`redteam_agent`	Offensive security testing
`reconnaissance`	OSINT & network scanning
`boot2root`	Full exploitation chains
`flag_discriminator`	CTF flag validation

Switch with: /agent [number]

Common Workflows

CTF Challenge

cai> Solve this web vulnerability challenge at <URL>

Network Reconnaissance

cai> Target IP: 192.168.x.x, perform a full network scan

Model / Cost Management

/model      # Switch LLM provider or model
/cost       # View token usage and spending
/mcp        # Manage MCP integrations
/load logs/session.jsonl   # Resume prior session

Environment Setup

Create .env in working directory:

ANTHROPIC_API_KEY=sk-ant-...
OPENAI_API_KEY=sk-...        # optional
CAI_LICENSE_OFF=1            # skip license for open usage
CAI_TRACING=true             # enable OpenTelemetry tracing

Install / Update

pip install cai-framework
pip install --upgrade cai-framework

Source: https://github.com/aliasrobotics/cai

---name: caidescription: Launch and interact with CAI (Cybersecurity AI) — an AI-powered offensive/defensive security framework with 300+ model support. Use for CTF challenges, penetration testing workflows, reconnaissance, and vulnerability research.---# CAI — Cybersecurity AICAI is an open-source AI-powered security framework by Alias Robotics.- **300+ AI models** (Anthropic, OpenAI, DeepSeek, Ollama, etc.)- **Specialized agents**: CTF, recon, red team, boot2root, flag discriminator- **Human-in-the-loop** via Ctrl+C intervention## Quick Launch```bash# Verify installationpip show cai-framework# Launch interactive shell (binary path on macOS with system Python 3.9)/Users/tananan/Library/Python/3.9/bin/cai# Or add to PATH permanently:# echo 'export PATH="$HOME/Library/Python/3.9/bin:$PATH"' >> ~/.zshrc && source ~/.zshrc# Then just: cai# Disable license check if neededCAI_LICENSE_OFF=1 /Users/tananan/Library/Python/3.9/bin/cai```## Agent SelectionOnce inside `cai`, use `/agent` to list and switch agents:| Agent | Purpose ||-------|---------|| `ctf_agent` | CTF challenge solving || `redteam_agent` | Offensive security testing || `reconnaissance` | OSINT & network scanning || `boot2root` | Full exploitation chains || `flag_discriminator` | CTF flag validation |Switch with: `/agent [number]`## Common Workflows### CTF Challenge```cai> Solve this web vulnerability challenge at <URL>```### Network Reconnaissance```cai> Target IP: 192.168.x.x, perform a full network scan```### Model / Cost Management```/model # Switch LLM provider or model/cost # View token usage and spending/mcp # Manage MCP integrations/load logs/session.jsonl # Resume prior session```## Environment SetupCreate `.env` in working directory:```envANTHROPIC_API_KEY=sk-ant-...OPENAI_API_KEY=sk-... # optionalCAI_LICENSE_OFF=1 # skip license for open usageCAI_TRACING=true # enable OpenTelemetry tracing```## Install / Update```bashpip install cai-frameworkpip install --upgrade cai-framework```Source: https://github.com/aliasrobotics/cai

Frequently asked questions

Cai