protect-mcp-setup

Enforce Cedar policy compliance and generate Ed25519-signed receipts for Claude Code tool calls.

Updated Jul 8, 2026
One-click install
npx skills add https://github.com/PriyanshKuniyal/gemini-cli-resources --skill protect-mcp-setup-priyanshkuniyal
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: protect-mcp-setup
Source: https://github.com/PriyanshKuniyal/gemini-cli-resources/tree/main/extensions/claude-code-workflows/plugins/protect-mcp/skills/protect-mcp-setup
Command: npx skills add https://github.com/PriyanshKuniyal/gemini-cli-resources --skill protect-mcp-setup-priyanshkuniyal

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires npx, claude, cedar-policy, and includes scripts (resource) components.

What problem does it solve?

This Skill addresses the need for secure, policy-compliant, and auditable tool calls within Claude Code projects.

Core Features & Use Cases

  • Policy Enforcement: Ensures all tool calls are compliant with predefined policies using AWS's Cedar engine.
  • Ed25519-Signed Receipts: Provides tamper-evident evidence for every tool call's input, decision, and outcome.
  • Offline Verification: Allows receipt verification without an internet connection, enhancing security and compliance.
  • Use Case: A developer can add protect-mcp to a Claude Code project to enforce strict policy checks on tool usage, such as 'rm -rf' or editing project files outside of the project root, while ensuring all actions are logged and can be verified by third parties.

Quick Start

Set up protect-mcp in your Claude Code project by following these steps:

claude plugin install wshobson/agents/protect-mcp
npx protect-mcp@latest serve --enforce

Frequently Asked Questions about protect-mcp-setup

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I enforce policy compliance for Claude Code tool calls?

You can enforce policy compliance for Claude Code tool calls by installing the protect-mcp plugin, which utilizes AWS's Cedar engine to ensure all tool actions adhere to predefined security policies before execution.

What are Ed25519-signed receipts used for in tool usage security?

Ed25519-signed receipts provide tamper-evident evidence for every tool call's input, decision, and outcome, allowing third parties to verify logged actions and ensure strict tool usage security.

Can I verify Claude Code tool call logs without an internet connection?

Yes, you can verify Claude Code tool call logs offline using Ed25519 signature generation, which provides tamper-evident receipts that allow third-party receipt verification without requiring an internet connection.

How do I prevent unsafe commands like rm -rf in Claude Code projects?

To prevent unsafe commands like rm -rf in Claude Code projects, apply Cedar policy enforcement via the protect-mcp plugin to strictly check and block tool usage that violates your defined project policies.

Do I need Cedar policy engine to secure Claude Code tool actions?

Yes, Cedar policy engine is a required dependency for securing Claude Code tool actions, as it provides the core policy enforcement mechanism that validates tool inputs and outcomes against your rules.

What is the best way to add auditable actions to Claude Code projects?

The best way to add auditable actions to Claude Code projects is using the protect-mcp plugin with Cedar policy enforcement, generating tamper-evident Ed25519-signed receipts for every executed tool call.