菜单
List, triage, and monitor vulnerabilities detected across your AI agents. Shows severity ratings, categories, status, and remediation recommendations. Use when checking security posture or reviewing findings.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Authenticate with Audn.ai for red-team security testing. Sets up OAuth login or API token. Run this before using any other audn skill. Use when the user needs to log in, authenticate, connect to Audn, or set up their API credentials.
Register and verify an AI agent on Audn.ai for security testing. Supports text API agents (OpenAI, Anthropic, custom LLMs) and voice agents (Twilio, Genesys, Amazon Connect). Use when the user wants to add a new agent to test.
Generate ISO 42001, SOC2, vulnerability, executive, or compliance reports from Audn.ai campaign results. Use when the user needs compliance evidence, audit documentation, or executive summaries of security testing.
Quick vulnerability scan of an AI agent endpoint. Runs only critical and high-severity attack scenarios for a fast security check. Use for a rapid smoke test before a full red-team audit.
Run a full adversarial red-team security audit against any AI agent (text or voice). Registers the agent, selects attack scenarios, executes a campaign, monitors progress, retrieves results, and generates a vulnerability report. Use when the user wants to security-test, red-team, pen-test, or audit an LLM or voice agent.
Create custom adversarial attack scenarios for AI agent red-teaming. Supports prompt injection, jailbreak, data extraction, social engineering, emotional manipulation, and role confusion attacks. Use when the user wants to design a specific attack test.
| name | audn-vuln-monitor |
| description | List, triage, and monitor vulnerabilities detected across your AI agents. Shows severity ratings, categories, status, and remediation recommendations. Use when checking security posture or reviewing findings. |
View and triage vulnerabilities detected by Audn.ai security campaigns.
Bearer token via AUDN_API_TOKEN env var. Base URL: https://audn.ai.
All requests include header: Authorization: Bearer $AUDN_API_TOKEN.
GET /api/vulnerabilities
Returns an array of findings with: id, title, severity (critical/high/medium/low/info), status (open/in_progress/resolved/suppressed), category, description, detected_at, recommendations.
For deeper context:
GET /api/results?limit=20&status=fail&sortOrder=desc
GET /api/results/<result_id>
Returns full transcript, analysis, attack details, and linked campaign/agent/scenario.
GET /api/campaigns/<campaign_id>
## Vulnerability Dashboard
**Total**: X findings | **Critical**: A | **High**: B | **Medium**: C | **Low**: D
### Open Vulnerabilities
| # | Severity | Category | Title | Agent | Detected |
|---|----------|----------|-------|-------|----------|
| 1 | CRITICAL | Prompt Injection | System prompt extractable via... | GPT-4 Bot | 2026-04-10 |
| 2 | HIGH | Data Extraction | PII leakage through indirect query | Claude Assistant | 2026-04-09 |
| 3 | MEDIUM | Social Engineering | Partial compliance with authority claim | Internal Bot | 2026-04-08 |
### Remediation Priority
1. **[CRITICAL]** <title> — <recommendation>
2. **[HIGH]** <title> — <recommendation>
## Vulnerability Detail — <title>
**Severity**: CRITICAL
**Category**: Prompt Injection
**Status**: open
**Detected**: <date>
**Agent**: <agent name>
### Description
<full description>
### Recommendations
1. <recommendation 1>
2. <recommendation 2>
### Related Test Result
**Campaign**: <campaign name>
**Scenario**: <scenario name>
**Result**: FAIL
**Transcript excerpt**: <relevant portion of conversation>
If user asks about trends, fetch multiple data points:
GET /api/results?limit=100&sortBy=created_at&sortOrder=desc
Group by week/category and report:
$ARGUMENTS