Conduct periodic access reviews and certifications. Implement access governance and recertification workflows. Use when managing access compliance.
日本語の概要は準備中です。原文の説明を表示しています。
Perform basic penetration testing and security assessments. Use reconnaissance, vulnerability discovery, and exploitation techniques. Use when validating security controls or assessing system security.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Validate security controls through authorized testing.
pentest_phases:
1_reconnaissance:
- Passive information gathering
- DNS enumeration
- Network mapping
2_scanning:
- Port scanning
- Service identification
- Vulnerability scanning
3_exploitation:
- Attempt exploitation
- Verify vulnerabilities
- Document findings
4_post_exploitation:
- Privilege escalation
- Lateral movement
- Data access
5_reporting:
- Document findings
- Risk assessment
- Remediation recommendations
# DNS enumeration
dig example.com ANY
host -l example.com
# Subdomain discovery
subfinder -d example.com
# WHOIS
whois example.com
# Port scan
nmap -sV -sC -p- target.com
# Web scanning
nikto -h https://target.com
dirb https://target.com
# Vulnerability scan
nmap --script vuln target.com
# SQL injection test
sqlmap -u "http://target.com/page?id=1"
# XSS testing
# Use Burp Suite or manual testing
# Directory traversal
curl "http://target.com/file?path=../../../etc/passwd"
scope:
in_scope:
- target.com
- api.target.com
out_of_scope:
- production-db.target.com
- third-party services
testing_window: "Weekdays 2-6 AM UTC"
emergency_contact: "security@target.com"
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Conduct periodic access reviews and certifications. Implement access governance and recertification workflows. Use when managing access compliance.
日本語の概要は準備中です。原文の説明を表示しています。
Build automated evaluation suites for AI agents using golden datasets, rubrics, and regression gates. Use when shipping agent features, validating prompt changes, or gating deployments on quality.
日本語の概要は準備中です。原文の説明を表示しています。
Instrument AI agents with tracing, token metrics, latency, and cost visibility. Use for reliability and debugging.
日本語の概要は準備中です。原文の説明を表示しています。
Secure AI agents against prompt injection, tool abuse, and data exfiltration with defense-in-depth controls. Use when building, deploying, or hardening agentic AI systems that invoke tools, access data, or interact with production infrastructure.
日本語の概要は準備中です。原文の説明を表示しています。
Secure AI coding agents (Claude Code, Cursor, Codex, Copilot) with permission boundaries, secret protection, code review gates, and safe sandbox configurations for team environments.
日本語の概要は準備中です。原文の説明を表示しています。
Use service mesh patterns for AI inference traffic management, mTLS, canary releases, policy enforcement, and cross-cluster resilience.
日本語の概要は準備中です。原文の説明を表示しています。