cis-benchmarks – skill Bảo mật cho agent AI
Kiểm tra và khắc phục các vi phạm benchmark CIS.
Cài skill cis-benchmarks bằng một lệnh
Chọn công cụ bạn đang dùng, copy lệnh và dán vào terminal. Skill sẽ nằm trong thư mục skills của công cụ đó và sẵn sàng dùng ở phiên làm việc tiếp theo.
Claude Code
mkdir -p ~/.claude/skills && curl -fsSL https://skill.dn1nguoi.com/z/cis-benchmarks-14128.tgz | tar xz -C ~/.claude/skills
Gọi trong Claude Code: /cis-benchmarks – rồi mô tả công việc bạn muốn làm.
Codex
mkdir -p ~/.codex/skills && curl -fsSL https://skill.dn1nguoi.com/z/cis-benchmarks-14128.tgz | tar xz -C ~/.codex/skills
Gọi trong Codex: $cis-benchmarks – rồi mô tả công việc bạn muốn làm.
Hermes Agent
mkdir -p ~/.hermes/skills && curl -fsSL https://skill.dn1nguoi.com/z/cis-benchmarks-14128.tgz | tar xz -C ~/.hermes/skills
Gọi trong Hermes Agent: /cis-benchmarks – rồi mô tả công việc bạn muốn làm.
DeepSeek
mkdir -p ~/.deepcode/skills && curl -fsSL https://skill.dn1nguoi.com/z/cis-benchmarks-14128.tgz | tar xz -C ~/.deepcode/skills
Gọi trong DeepSeek: /cis-benchmarks – rồi mô tả công việc bạn muốn làm.
Nội dung SKILL.md
Nội dung SKILL.md gốc bằng tiếng Anh.
CIS Benchmarks
Assess a selected benchmark/version against explicitly authorized systems. Scanner output is partial evidence, not certification; verify applicability, documented exceptions and operational impact before remediation.
When to Use This Skill
Use this skill when:
- Assessing security compliance
- Implementing security baselines
- Meeting regulatory requirements
- Hardening systems to standards
Assessment Tools
OpenSCAP
# Install
apt install openscap-scanner scap-security-guide
# Confirm installed content matches the actual OS and required benchmark/version.
SCAP_FILE="<approved-installed-datastream.xml>"
oscap info "$SCAP_FILE"
PROFILE_ID="<exact-CIS-profile-listed-by-oscap-info>"
# Run the selected assessment only after scope and report-location approval.
oscap xccdf eval \
--profile "$PROFILE_ID" \
--results results.xml \
--report report.html \
"$SCAP_FILE"Lynis
# Install
apt install lynis
# Run audit
lynis audit system
# Generate report
lynis audit system --report-file "$PRIVATE_AUDIT_DIR/lynis-report.dat"InSpec
# cis-profile/controls/ssh.rb
control 'cis-ssh-1' do
impact 1.0
title 'Ensure SSH root login is disabled'
describe sshd_config do
its('PermitRootLogin') { should eq 'no' }
end
end
control 'cis-ssh-2' do
impact 0.7
title 'Ensure SSH password authentication is disabled'
describe sshd_config do
its('PasswordAuthentication') { should eq 'no' }
end
end# Run InSpec
inspec exec cis-profile -t ssh://user@targetKubernetes CIS
# kube-bench
docker run --rm -v /etc:/etc:ro -v /var:/var:ro \
aquasec/kube-bench@sha256:<reviewed-image-digest> run --targets node
# Check specific sections
kube-bench --help
# Select role/sections supported by the reviewed version and actual cluster; do not assume master is a current target name.Remediation Workflow
workflow:
1_scan:
- Run automated assessment
- Generate baseline report
2_analyze:
- Review findings
- Identify false positives
- Prioritize by risk
3_remediate:
- Apply fixes
- Document exceptions
- Verify changes
4_validate:
- Re-run assessment
- Confirm remediation
- Generate compliance reportBest Practices
- Baseline before hardening
- Document exceptions
- Automate assessments
- Track compliance over time
- Regular re-assessment
- Version control configurations
Related Skills
- linux-hardening (
linux-hardening) - Linux security - vulnerability-scanning (
vulnerability-scanning) - Security scanning
Limitations
- An OpenSCAP/Lynis assessment may require privileges and read sensitive host configuration; approve the exact host, profile, privileges, private outputs and limits first. Do not enable auto-remediation or mount all host paths without reviewing necessity. A generic Lynis audit is not a complete CIS benchmark.
- Benchmark/profile identifiers and installed SCAP content vary; use [OpenSCAP’s documented info/eval flow](https://static.open-scap.org/openscap-1.4.1/oscap_user_manual.html). Test changes and preserve rollback/access before hardening production.
- Docs-only import: upstream scripts and templates not bundled.
Example
# Read-only first: inventory before any active step.
command -v oscap
oscap --version> Adapted from [BagelHole/DevOps-Security-Agent-Skills](https://github.com/BagelHole/DevOps-Security-Agent-Skills) (MIT); frontmatter, When to Use/Limitations, and safety boundaries added for upstream compliance. Docs-only import: helper scripts and templates not bundled.
Skill liên quan
- incident-response – Dùng khi một sự cố bảo mật đã được phát hiện hoặc khai báo, cần phân loại, sàng lọc ưu…
- information-security-manager-iso27001 – Triển khai ISMS theo ISO 27001 và quản trị an ninh mạng cho các công ty HealthTech và…
- ops-proxyns – Bảo vệ proxy ở mức kernel qua proxy-ns — buộc TOÀN BỘ lưu lượng (TCP/UDP/DNS) đi qua Tor…
- claude-settings-audit – Phân tích repository để tạo quyền settings.json khuyến nghị cho Claude Code. Dùng khi…
- origin-ip-discovery – Tìm IP gốc phía sau CDN/WAF qua hash favicon, lịch sử DNS và chứng chỉ SSL.
- client-secret-exposure-audit – Kiểm toán ứng dụng web đã triển khai để tìm secret bị lộ ra trình duyệt: API key/token…
- osint-methodology – Phương pháp OSINT toàn diện cho chiến dịch red-team bên ngoài và đánh giá bề mặt tấn…
- cloudflare-security-audit – Kiểm tra codebase được ủy quyền để tìm lỗ hổng có thể khai thác, qua trinh sát có phạm…