1. Conduid
  2. Security
  3. Guardrail
MCP server · Security

Guardrail

by · github.com/klars-ai/guardrail

A firewall for AI agent tool calls. Wrap any MCP server in one command and block dangerous actions before they execute.

Unclaimed ai-agentsclaudeguardrailsllmmcpmodel-context-protocolsecurity
InstallView source
37Low

Scored 4 hours ago · breakdown

About Guardrail

Guardrail is an MCP server published by klars-ai in the Security category: a firewall for AI agent tool calls. Wrap any MCP server in one command and block dangerous actions before they execute.

Recent commit activity is the strongest signal that reported issues will get attention.

Install

Claude Code
claude mcp add guardrail-7b3a -- npx -y @klars/guardrail
npx
npx -y @klars/guardrail

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Guardrail

Powered by Claude · Grounded in docs

I know everything about Guardrail. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

6 checks have not been run yet.

Questions

About Guardrail

How do I install Guardrail?

Run claude mcp add guardrail-7b3a -- npx -y @klars/guardrail, then add the server to your MCP client's configuration.

Is Guardrail safe to use with an AI agent?

Its trust score is 37 out of 100 (low). Conduid hasn't run static security checks on this repository yet, so review the source yourself before granting it credentials. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Guardrail still maintained?

Yes — the last commit was a month ago. The repository has 0 stars and 0 forks.