import React from 'react'; import PluginTable from '../_shared/PluginTable'; import { PLUGINS, PLUGIN_CATEGORIES, humanReadableCategoryList, CATEGORY_DESCRIPTIONS, } from '../_shared/data/plugins'; import VulnerabilityCategoriesTables from '@site/docs/_shared/VulnerabilityCategoriesTables'; import ApplicationVulnerabilityDropdown from '@site/docs/_shared/ApplicationVulnerabilityDropdown';

Types of LLM vulnerabilities

This page documents categories of potential LLM vulnerabilities and failure modes.

Each vulnerability type is supported Promptfoo's open-source plugins. Plugins are a modular system for testing risks and vulnerabilities in LLM models and applications. See the quickstart guide to run your first red team.

See also our specific guides on:

• Red teaming AI agents
• Red teaming RAGs
• Red teaming multi-modal models
• Testing and validating guardrails

:::note Interpreting Attack Success Rates When comparing red team results across different tools or papers, be aware that Attack Success Rate (ASR) depends heavily on attempt budget, prompt set composition, and judge choice. See Why ASR Isn't Comparable Across Jailbreak Papers for guidance on interpreting these metrics. :::

Vulnerability Types

Security Vulnerabilities

Privacy Vulnerabilities

Criminal Activity

Harmful Activity

Misinformation and Misuse

Bias

Ecommerce

Financial

Medical

Pharmacy

Insurance

Custom

Vulnerabilities by Application

Not all applications are vulnerable to certain types of exploits. Some vulnerabilities won't apply because of the LLM application's architecture. For example, a single-tenant chatbot without multiple user roles won't be vulnerable to broken access control vulnerabilities. Select a category below to see where vulnerabilities may not apply.

Plugin Reference

For a complete list of available plugins and their severity levels, see the Plugins Overview page.