# mcp-seatbelt

> mcp-seatbelt — kryptosai-mcp-seatbelt. Use this tool when you need to protect AI agents from executing hazardous operations by blocking dangerous tool calls at runtime. It solves the problem of preventing unintended or malicious actions by enforcing a default-deny policy proxy. The mcp-seatbelt tool takes in AI agent tool calls as input and outputs a safe and controlled execution environment.

Canonical page: https://skillsregistry.net/skills/kryptosai-mcp-seatbelt  
JSON: https://api.skillsregistry.net/v1/skills/kryptosai-mcp-seatbelt

## Description

Runtime guardrails for AI agent MCP tools. Blocks dangerous tool calls at runtime with a default-deny policy proxy.

## Trust

- **Trust score (0–1):** 0.68
- **Verification tier:** scanned
- **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/KryptosAI/mcp-seatbelt)

## 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": "kryptosai-mcp-seatbelt"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/kryptosai-mcp-seatbelt` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/kryptosai-mcp-seatbelt/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