# pod

> pod — suhui-organization-pod. Use this tool when you need to compile and verify the actions of AI agents with least-privilege access, ensuring security and compliance by generating policies based on actual agent behavior. It takes in agent activity logs as input and outputs a provable policy, allowing for auditing and validation of agent actions. Ideal for use cases requiring strict access control and transparency, such as secure git operations.

Canonical page: https://skillsregistry.net/skills/suhui-organization-pod  
JSON: https://api.skillsregistry.net/v1/skills/suhui-organization-pod

## Description

pod — least-privilege compiler for AI agents. Compile a least-privilege policy from what your agent actually did, then prove what happened.

## Trust

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

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** container
- **Runtime environment:** vm
- **License:** Apache-2.0
- **Updated:** 2026-09-22

## Source

- **Source listing:** [GitHub](https://github.com/suhui-organization/pod)

## 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": "suhui-organization-pod"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/suhui-organization-pod` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/suhui-organization-pod/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