# vigil

> vigil — robinoppenstam-vigil. Use this tool when you need to ensure safe and controlled interactions between AI agents and external tools, preventing potential errors or security breaches. Vigil provides guardrails for tool calls, solving problems related to unauthorized access, data leaks, or malicious activities. It takes in tool call requests as input and outputs validated, secure calls, making it ideal for use in high-stakes or sensitive applications.

Canonical page: https://skillsregistry.net/skills/robinoppenstam-vigil  
JSON: https://api.skillsregistry.net/v1/skills/robinoppenstam-vigil

## Description

AI agent safety guardrails for tool calls.

## Trust

- **Trust score (0–1):** 0.65
- **Verification tier:** scanned
- **Last scanned:** 2026-05-20

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** instructions
- **Runtime environment:** llm
- **Category:** ai-ml
- **Updated:** 2026-09-16

## Source

- **Source listing:** [ClawHub](https://clawskills.sh/skills/robinoppenstam-vigil)

## 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": "robinoppenstam-vigil"
    }
  }
}
```

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