Skip to content
Agentmart
Submit a tool

Agent skills for guardrails and AI safety in Claude Code

287 agent skills for guardrails and AI safety in the catalog list Claude Code as a supported host. These are the 15 that rank highest, ordered by adoption, freshness and documentation. ai-coding-agent-guardrails, git-guardrails-claude-code and ai-red-teaming lead the list. Claude Code reads them from .claude/skills/<name>/SKILL.md in the project, or ~/.claude/skills/<name>/ for every project. A skill is a folder with a SKILL.md file; Claude Code loads it when a task matches its description. Every listing links to its source and shows a freshness grade.

Updated Oct 26, 2026 · Built from catalog data; how we rank and grade

How to install in Claude Code

npx skills add <owner>/<repo> --skill <name> -a claude-code

Where it goes: .claude/skills/<name>/SKILL.md in the project, or ~/.claude/skills/<name>/ for every project.

The host's docs describe copying a folder with a SKILL.md into a skills directory; the npx skills command is a third-party shortcut for the same thing.

Each listing page has the exact command for that listing.

Top 15 of 287

  1. 1

    Secure AI coding agents (Claude Code, Cursor, Codex, Copilot) with permission boundaries, secret protection, code review gates, and safe sandbox…

    • Repository has 48.4K stars
    • Active in the last 30 days
    • Open source (MIT)

    ★ 48.4KLast active Oct 24, 2026Claude Code, Claude apps, Codex, Cursor +6

  2. 2

    设置 Claude Code hooks,在危险 git commands(push、reset --hard、clean、branch -D 等)执行前阻止它们。适用于用户想防止破坏性 git 操作、添加 git safety hooks,或在 Claude Code 中阻止 git push/reset 时。

    • Repository has 5.2K stars
    • Active in the last 30 days
    • Open source (MIT)

    ★ 5.2KLast active Oct 8, 2026Claude Code, Claude apps, Codex, Cursor +6

  3. 3
    ai-red-teaming

    devops-skills

    Run structured AI red team exercises for jailbreak resistance, data exfiltration risk, harmful output controls, and agent tool abuse resilience.

    • Repository has 48.4K stars
    • Active in the last 30 days
    • Open source (MIT)

    ★ 48.4KLast active Oct 24, 2026Claude Code, Claude apps, Codex, Cursor +6

  4. 4

    设置 Claude Code 钩子,在危险 Git 命令(push、reset --hard、clean、branch -D 等)执行前将其拦截。当用户想要防止破坏性 Git 操作、添加 Git 安全钩子或在 Claude Code 中阻止 git push/reset 时使用。

    • Active in the last 30 days
    • Open source (MIT)
    • Ships helper scripts

    ★ 507Last active Oct 20, 2026Claude Code, Claude apps, Codex, Cursor +6

  5. 5
    aig-scanner

    TencentOfficial

    A.I.G Scanner — AI security scanning for infrastructure, AI tools / skills, AI Agents, and LLM jailbreak evaluation via Tencent Zhuque Lab AI-Infra-Guard.

    • Published by the vendor
    • Repository has 7.1K stars
    • Active in the last 30 days

    ★ 7.1KLast active Oct 22, 2026Claude Code, Claude apps, Codex, Cursor +6

  6. 6

    Best practices for building agents and agentic applications with Plano, including configuration, routing, orchestration, guardrails, observability, and…

    • Repository has 7.1K stars
    • Active in the last 30 days
    • Open source (Apache-2.0)

    ★ 7.1KLast active Oct 24, 2026Claude Code, Claude apps, Codex, Cursor +6

  7. 7

    Harden Plano filter chains and guardrails.

    • Repository has 7.1K stars
    • Active in the last 30 days
    • Open source (Apache-2.0)

    ★ 7.1KLast active Oct 24, 2026Claude Code, Claude apps, Codex, Cursor +6

  8. 8
    paid-ads-amazon

    nowork-studio

    Plan and review Amazon Ads with margin-aware ACoS, product, and search-term guardrails.

    • Repository has 4.1K stars
    • Active in the last 30 days
    • Open source (MIT)

    ★ 4.1KLast active Oct 9, 2026Claude Code, Claude apps, Codex, Cursor +6

  9. 9
    careful

    garrytan

    Safety guardrails for destructive commands.

    • Repository has 137.5K stars
    • Active in the last 30 days
    • Open source (MIT)

    ★ 137.5KLast active Oct 19, 2026Claude Code, Claude apps, Codex, Cursor +6

  10. 10

    Configure the shared guard against catastrophic shell commands in local AI agents.

    • Repository has 4.2K stars
    • Active in the last 30 days
    • Open source (MIT)

    ★ 4.2KLast active Oct 18, 2026Claude Code, Claude apps, Codex, Cursor +6

  11. 11

    iOS IPA static analysis — hardcoded secrets, ATS bypass, Keychain, jailbreak detection, reverse engineering with class-dump/ipsw.

    • Active in the last 30 days
    • Open source (MIT)

    ★ 883Last active Oct 23, 2026Claude Code, Claude apps, Codex, Cursor +6

  12. 12

    Set up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute.

    • Largest repository here (302,383 stars)
    • Active in the last 30 days
    • Open source (MIT)

    ★ 302.4KLast active Oct 4, 2026Claude Code, Claude apps, Codex, Cursor +6

  13. 13
    investigation-cost-guardrail

    tqquresh, inesttiaOfficial

    Cost guardrail for AWS DevOps Agent that covers ALL AWS services and native agent tools.

    • Published by the vendor
    • Active in the last 30 days
    • Open source (Apache-2.0)

    ★ 116Last active Oct 21, 2026Claude Code, Claude apps, Codex, Cursor +6

  14. 14
    nemotron-policy-generator

    Shyamala Prayaga <sprayaga@nvidia.com>

    Generates BYO custom safety policies for NVIDIA Nemotron content-safety guardrails — Nemotron-Content-Safety-Reasoning-4B (text) and multimodal…

    • Repository has 3.7K stars
    • Open source (Apache-2.0 AND CC-BY-4.0)

    ★ 3.7KLast active Jun 4, 2026Claude Code, Claude apps, Codex, Cursor +6

  15. 15
    engram-architecture-guardrails

    gentleman-programming

    Architecture guardrails for Engram across local store, cloud sync, dashboard, and plugins.

    • Repository has 7.6K stars
    • Active in the last 30 days
    • Open source (Apache-2.0)

    ★ 7.6KLast active Oct 23, 2026Claude Code, Claude apps, Codex, Cursor +6