protect-mcp

protect-mcp is a skill for Claude Code from ScopeBlind/scopeblind-gateway. It costs 40 tokens per session (631 once invoked), scanned C, original, from a forked repository, MIT.

A security layer for AI-agent tool calls that checks rules and creates signed records. Cedar is a language for writing allow, deny, or review policies; Ed25519 is a cryptographic signature method that can be checked offline.

In plain words
What is it for?
It is used to enforce tool-call policies and create or verify signed receipts for individual calls or a complete activity log.
Why use it?
It helps control which tool calls are permitted and provides tamper-evident evidence of what the agent did.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions Claude Code.

Part of the protect-mcp plugin — 1 skill, 2 hooks, 1 MCP server shipped together

Good fit It is used to enforce tool-call policies and create or verify signed receipts for individual calls or a complete activity log.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/scopeblind/scopeblind-gateway/protect-mcp
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add ScopeBlind/scopeblind-gateway --skill protect-mcp
Clone the repo
git clone --depth 1 https://github.com/ScopeBlind/scopeblind-gateway

Made for: Claude Code.

Or install protect-mcp, the plugin that ships this one along with the rest of its 1 skill, 2 hooks, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for protect-mcp

README.md
[![agentmods](https://agentmods.dev/badge/skills/scopeblind/scopeblind-gateway/protect-mcp/github.svg)](https://agentmods.dev/skills/scopeblind/scopeblind-gateway/protect-mcp)
Your own site
<a href="https://agentmods.dev/skills/scopeblind/scopeblind-gateway/protect-mcp"><img src="https://agentmods.dev/badge/skills/scopeblind/scopeblind-gateway/protect-mcp/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for protect-mcp

Your own site · 80×15
<a href="https://agentmods.dev/skills/scopeblind/scopeblind-gateway/protect-mcp"><img src="https://agentmods.dev/badge/skills/scopeblind/scopeblind-gateway/protect-mcp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 40 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 631 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin fork From a forked repository.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00040 $0.00631
Opus 5 $0.00020 $0.00316
Sonnet 5 $0.00008 $0.00126
Haiku 4.5 $0.00004 $0.00063

Measured 11d ago against content hash 3982a6ea0787, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade C, and why

protect-mcp scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

context.input.command like "rm -rf *"
skills/protect-mcp/SKILL.md · 89 lines

What it actually says

protect-mcp — Agent Governance with Cryptographic Receipts

What This Does

Adds two layers to every tool call:

  1. Cedar policy enforcement — evaluate allow/deny decisions using the same authorization engine AWS uses for IAM. Policies are declarative .cedar files.
  2. Ed25519 receipt signing — each tool call produces a tamper-evident receipt. If the record is modified after signing, the signature breaks. Anyone can verify offline without trusting the issuer.

Setup

# One-time: configure Claude Code hooks
npx [email protected] init-hooks

# Start the hook server (runs on port 9377)
npx [email protected] serve --enforce

This configures Claude Code to POST every tool call event to protect-mcp for policy evaluation and receipt signing. First run auto-generates permissive Cedar policies that you can tighten.

Usage

Once running, protect-mcp operates silently. Every tool call:

  • Is evaluated against Cedar policies (allow/deny/ask)
  • Produces a signed receipt in .protect-mcp-log.jsonl
  • Decision is returned to Claude Code within ~1ms

Verify Receipts

# Verify all receipts in the log
npx @veritasacta/verify .protect-mcp-log.jsonl

# Verify a single receipt
npx @veritasacta/verify receipt.json

What a Receipt Looks Like

{
  "type": "protectmcp:decision",
  "tool_name": "Bash",
  "decision": "allow",
  "policy_digest": "sha256:a3f8...",
  "issued_at": "2026-04-04T12:00:00Z",
  "signature": { "alg": "EdDSA", "sig": "..." }
}

Cedar Policy Example

// Block destructive commands
forbid (
  principal,
  action == Action::"tool_call",
  resource == Tool::"Bash"
) when {
  context.input.command like "rm -rf *"
};
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 11d ago First seen · 89 lines · 40 tokens per session scan C 3982a6ea0787

Subscribe to this mod's changes

protect-mcp is a skill published in the GitHub repository ScopeBlind/scopeblind-gateway (9 stars, last pushed today), licensed MIT. It adds 40 tokens to every session and 631 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). It comes from a forked repository.

Related

Other skills, from other repositories

cis-aws-database-4.1

Ensure AWS Identity and Access Management (IAM) is in use.

CyberStrikeus/CyberStrike · 22 tokens

iam-review

Reviews identity and access management configurations against NIST SP 800-63B, NIST SP 800-207 zero trust principles, and CIS Controls v8. Auto-invoked when reviewing IAM policies, role definitions, user provisioning workflows, or when asked to assess identity security posture. Produces findings on least privilege…

UnitOneAI/SecuritySkills · 83 tokens

api-31-idor

IDOR (Insecure Direct Object Reference) testing checklist: object ID enumeration, horizontal/vertical privilege escalation, GUID predictability, indirect references via hashes, chained IDOR, and API endpoint IDOR. Use for web app pentests and bug bounty IDOR discovery.

vigilantshield/Claude-HunterKit · 60 tokens

api-04-bola

Broken Object Level Authorization (BOLA) testing: object ID enumeration in REST/GraphQL/gRPC, horizontal/vertical privilege escalation, UUID prediction, mass assignment via extra parameters, and indirect reference manipulation. API-specific variant of IDOR. Use when testing API authorization.

vigilantshield/Claude-HunterKit · 60 tokens

api-07-bfla

Broken Function Level Authorization (BFLA) testing: horizontal/vertical function access, admin endpoint enumeration, HTTP method bypass, role confusion via header manipulation, and GraphQL mutation access control. Use when testing API authorization for privileged functions.

vigilantshield/Claude-HunterKit · 54 tokens

api-08-advanced-bola

Advanced BOLA/IDOR techniques: batch/parallel enumeration, side-channel detection via response timing, CDN cache-key probing, gRPC/protobuf ID manipulation, and bulk data export abuse. Use when basic BOLA tests are blocked by WAF or rate limits.

vigilantshield/Claude-HunterKit · 61 tokens