protect-mcp-setup

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

1|Updated Apr 27, 2026
One-click install
npx skills add https://github.com/haxlys/skills --skill protect-mcp-setup-haxlys
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: protect-mcp-setup
Source: https://github.com/haxlys/skills/tree/main/vendored/wshobson-agents/plugins/protect-mcp/skills/protect-mcp-setup
Command: npx skills add https://github.com/haxlys/skills --skill protect-mcp-setup-haxlys

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Claude Code sessions often lack a trustworthy audit trail: tool calls can be executed without verifiable records, making compliance and accountability difficult.

Core Features & Use Cases

  • Policy evaluation hooks ensure only permitted tool usage by evaluating Cedar policies before execution.
  • Ed25519-signed receipts provide verifiable, offline-easily-auditable records of decisions and inputs.
  • Easy project integration with local signing servers for compliant tool governance across teams and environments.

Quick Start

Install protect-mcp, configure hooks in .claude/settings.json, and run the local signing server to start producing signed receipts.

Frequently Asked Questions about protect-mcp-setup

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

FAQPage Schema
How do I add cryptographic governance to Claude Code tool usage?

You can add cryptographic governance to Claude Code by installing protect-mcp, configuring hooks in .claude/settings.json, and running a local signing server to enforce Cedar policies and generate Ed25519-signed receipts.

How do offline verification and signed receipts work for compliance audit trails?

Offline verification of signed receipts uses Ed25519 cryptographic signatures to provide tamper-evident records of tool call decisions and inputs. This allows compliance audits to verify records without active network connections.

Can I use Cedar policies to enforce security governance before Claude Code executes tools?

Yes, policy evaluation hooks intercept Claude Code tool calls to evaluate Cedar policies before execution. This security governance mechanism ensures only explicitly permitted tool usage proceeds in development pipelines.

What is the best way to create tamper-evident audit trails for AI tool calls in development pipelines?

The best way to create tamper-evident audit trails is using Ed25519-signed receipts for every tool call decision. This cryptographic approach satisfies compliance requirements across development pipelines by providing offline verifiable records.

Do I need a local signing server to produce cryptographic audit receipts for Claude Code?

Yes, you need to run a local signing server to produce Ed25519-signed receipts. This server integrates with your project environment to issue verifiable records for policy evaluation decisions and tool inputs.

When do I need cryptographic audit trails for AI agent tool calls?

You need cryptographic audit trails when operating in compliance contexts or development pipelines requiring strict accountability. Tamper-evident records ensure tool calls can be verified offline and trusted across teams.