# agent-fence

> agent-fence — izumi6-agent-fence. Use this tool when you need to secure AI coding agents from potential threats, such as secret leaks and destructive commands. AgentFence solves problems related to unsafe Git operations and network egress by enforcing policy-as-code and MCP controls. It takes in AI agent actions as input and outputs allowed or blocked actions based on predefined security policies.

Canonical page: https://skillsregistry.net/skills/izumi6-agent-fence  
JSON: https://api.skillsregistry.net/v1/skills/izumi6-agent-fence

## Description

Local security layer for AI coding agents that blocks secret leaks, destructive commands, unsafe Git operations, and unauthorized network access.

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** container
- **Runtime environment:** vm
- **Updated:** 2026-09-24

## Source

- **Source listing:** [GitHub](https://github.com/Izumi6/agent-fence)

## 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": "izumi6-agent-fence"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/izumi6-agent-fence` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/izumi6-agent-fence/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