使い方
- インストールコマンドをコピーし、自分の環境へ導入する前にソースを確認します。
- 必要なツール、認証情報、対応環境を原文で確認します。
- 入力と成果物を指定し、出力を資料と照合します。
npx skills add https://github.com/smartworldarafath/rea_skill --skill strix開始前の確認
- 導入にはモデル契約、API枠、ファイルへのアクセス許可は含まれません。
- スター数や導入数は活動の指標であり、独立した品質・安全性試験ではありません。
見出し、例、コード、表、リンク、参照画像を含む原文を表示しています。
Strix AI Penetration Testing
Overview
This skill integrates [Strix](https://github.com/usestrix/strix) — the open-source AI penetration testing tool — into the REA_Skill agent ecosystem. Strix deploys autonomous AI hackers to find and fix your app's vulnerabilities, with seamless GitHub Actions / CI/CD integration to scan on every pull request and block insecure code before it reaches production.
Strix is strictly for targets you own or have explicit, written authorization to test. Unauthorized testing is illegal in most jurisdictions.
What Strix Does
- Autonomous AI hacking: AI agents that actively test your targets and propose fixes.
- CI/CD ready: Integrates with GitHub Actions to scan every pull request.
- Cloud scans:
strix cloud scans start --engagement-type live_test --domain-ids <uuid> --wait - Vulnerability listing:
strix cloud vulns list --severity critical(JSON output supported) - MCP servers: Connect your own Model Context Protocol servers via
~/.strix/mcp-servers.jsonand expose their tools to the agent during a run. - REST API: Every cloud operation has a matching
strix cloud <resource> <verb>command.
⚡ Integration with REA (Reverse Engineer Anything)
When combined with the REA suite:
- Pre-publish vulnerability hunt: Before shipping your app/website, run Strix to find weaknesses, then use REA skills to reverse-engineer and remediate the affected components.
- Web audit pivot: Use the Website Security Audit (
website-security-audit) skill for the 10-layer defensive scan, then Strix for active AI-driven penetration testing. - Binary ↔ App correlation: Recover source/assets with REA core skills, then point Strix at the running application to validate exploitability end-to-end.
# Example: start a live test scan and wait for results
strix cloud scans start --engagement-type live_test --domain-ids <uuid> --wait
# Example: list critical vulnerabilities as JSON
strix cloud vulns list --severity critical --json🚀 Setup & Installation
Strix is a Python/uv project. Install from the upstream repository:
# Clone Strix (upstream: https://github.com/usestrix/strix)
git clone https://github.com/usestrix/strix.git
cd strix
# Install dependencies
uv sync
# Run Strix
uv run strix --helpConfigure MCP servers (optional) in ~/.strix/mcp-servers.json:
[
{
"name": "github",
"transport": "http",
"url": "https://api.githubcopilot.com/mcp/",
"auth": { "kind": "bearer", "token": "your-token" },
"allowed_tools": ["list_issues"]
}
]📊 Result Handling
- Strix cloud JSON output is machine-readable; use
--jsonwhen stdout is not a terminal. - Binary downloads redirect raw bytes or use
--output FILE --jsonfor download metadata.
🛡️ Security & Red Lines
- Authorized use only: Run Strix exclusively against systems you own or have explicit, written permission to test, within the agreed scope.
- You are responsible for obtaining authorization and complying with applicable law.
- No third-party targets: Never point Strix at applications, websites, or APIs you do not own or are not explicitly authorized to assess.
- Provided "as is" with no warranty; misuse is your own responsibility.
See Also
- Strix repository: https://github.com/usestrix/strix
- Strix documentation: https://docs.strix.ai
- REASkill: https://github.com/smartworldarafath/REASkill