# mcpguard

> mcpguard — nbosa-mcpguard. Use this tool when you need to detect and prevent security threats in MCP environments, such as prompt injection, tool poisoning, and secrets exposure. It scans for potential vulnerabilities and provides a guard proxy for stdio, helping to identify and mitigate risks. Ideal for use cases involving git and MCP security scanning, where unsafe permissions and rug pulls need to be identified and addressed.

Canonical page: https://skillsregistry.net/skills/nbosa-mcpguard  
JSON: https://api.skillsregistry.net/v1/skills/nbosa-mcpguard

## Description

MCP security scanner and stdio guard proxy for detecting prompt injection, tool poisoning, rug pulls, secrets, and unsafe permissions.

## Trust

- **Trust score (0–1):** 0.93
- **Verification tier:** verified
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** container
- **Runtime environment:** vm
- **License:** MIT
- **Updated:** 2026-09-28

## Source

- **Source listing:** [GitHub](https://github.com/nbosa/mcpguard)

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "nbosa-mcpguard"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/nbosa-mcpguard` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/nbosa-mcpguard/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt