# agent-self-assessment

> Use this tool when you need to evaluate the security posture of AI agents, identifying vulnerabilities and weaknesses to ensure reliable operation. It solves problems related to agent security, compliance, and risk management by providing a comprehensive self-assessment framework. The tool takes in agent configurations and settings as input and outputs a detailed report highlighting potential security risks and recommendations for mitigation.

Canonical page: https://skillsregistry.net/skills/roosch269-agent-self-assessment  
JSON: https://api.skillsregistry.net/v1/skills/roosch269-agent-self-assessment

## Description

Security self-assessment tool for AI agents.

## Trust

- **Trust score (0–1):** 0.60
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** instructions
- **Runtime environment:** llm
- **Category:** ai-ml
- **Updated:** 2026-04-22

## Source

- **Source listing:** [ClawHub](https://clawskills.sh/skills/roosch269-agent-self-assessment)

## 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": "roosch269-agent-self-assessment"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/roosch269-agent-self-assessment` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/roosch269-agent-self-assessment/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