# Promptfoo

- Status: Unclaimed: compiled by Nowen from public sources, not reviewed by the company
- Industry: AI and Cybersecurity
- Delivery model: software
- Location: San Francisco, California, United States
- Website: https://promptfoo.dev
- Legal name: Promptfoo, Inc.
- AI agent profile: https://nowen.ai/agents/promptfoo-dev
- Updated: 2026-11-19
- Directory: one of 2 AI agents in AI and Cybersecurity (https://nowen.ai/industry/ai-and-cybersecurity)
- Read by AI assistants from: Perplexity

> Promptfoo helps developers and enterprises identify and address security risks in AI applications.

Promptfoo helps developers and enterprises build secure, reliable AI applications. It provides capabilities for identifying vulnerabilities, assessing risks, improving defenses, and supporting secure development workflows.

**Mission:** Securing the Future of AI

## Products & Services

### [Promptfoo Evaluations](https://www.promptfoo.dev/docs/getting-started/)
*Platform*
Evaluate prompts, models, and RAG pipelines to compare outputs, measure quality, and catch failures before they reach users.
Promptfoo Evaluations is an open-source CLI and library for testing prompts, models, RAG pipelines, and AI applications. Configure prompts, providers, test cases, and optional assertions, then run evaluations and review results in the terminal or web viewer. Use repeatable tests to compare outputs, assess quality, and identify failures.

- **Prompt and model comparisons** — Compare Outputs Across Prompts And Models
- **Automated output scoring** — Score Outputs Against Defined Requirements
- **Provider integrations** — Test Models Across 60+ Providers
- **RAG pipeline evaluation** — Evaluate Retrieval And Generated Answers
- **Local and CI/CD execution** — Run Evaluations Locally And In CI/CD
- **Evaluation setup and configuration** — Configure Evaluations Through CLI Or Web UI
- **Evaluation performance tools** — Speed Up Evaluation Workflows

### [Red Teaming](https://www.promptfoo.dev/red-teaming/)
*Solution*
Uncover application-specific AI vulnerabilities, track risk over time, and guide remediation with automated adversarial testing.
Promptfoo Red Teaming automatically tests AI applications, agents, and RAG systems with adversarial probes tailored to the target application. It covers more than 50 vulnerability types, including prompt injection, jailbreaks, data and PII leaks, harmful content, and unauthorized data access. Teams can connect targets through APIs, browser or direct model access, integrate tests with CI/CD or scheduled runs, and review findings with severity, test evidence, and remediation guidance. Findings can also be mapped to OWASP, NIST, the EU AI Act, and custom frameworks.

- **Application-specific attack generation** — Expose Weaknesses With Tailored Attacks
- **Vulnerability coverage** — Detect Risks Across 50+ Vulnerability Types
- **Vulnerability reports and remediation guidance** — Prioritize Findings And Guide Fixes
- **Continuous security testing** — Catch Regressions Across Releases
- **Custom attack strategies** — Adapt Attack Flows To Your Application
- **Target connectivity** — Test AI Targets Through Multiple Interfaces
- **Compliance mapping** — Map Findings To Security Frameworks
- **Flexible deployment** — Choose A Deployment That Fits Your Needs

### [Code Scanning](https://www.promptfoo.dev/code-scanning/)
*Product*
Find LLM-specific vulnerabilities early and give developers actionable fixes in their IDE, pull requests, and CI/CD workflows.
Promptfoo Code Scanning reviews code for vulnerabilities specific to applications built with LLMs and agents. Its AI agents trace data flows across repositories, including related files and modules, to identify issues such as prompt injection, data exfiltration, PII exposure, unsafe output handling, excessive agency, and jailbreak risks. Findings can appear as inline IDE diagnostics, pull request comments, or CI/CD output, with severity indicators, suggested fixes, and remediation prompts for AI coding agents. Teams can configure severity thresholds, tailor scanner sensitivity, and block risky pull requests or fail builds on critical findings.

- **LLM-specific vulnerability detection** — Find LLM-Specific Security Risks
- **Deep data-flow analysis** — Trace Risks Across Your Codebase
- **IDE scanning** — Catch Issues While Writing Code
- **Pull request security review** — Review And Block Risky Changes
- **CI/CD integration** — Gate Builds On Critical Findings
- **Actionable remediation guidance** — Guide Developers Toward Fixes
- **Configurable scanner sensitivity** — Tune Findings To Your Needs

### [Model Security](https://www.promptfoo.dev/model-security/)
*Solution*
Identify model-file risks, test behavior against attacks, compare model security, and document compliance to support safer deployment decisions.
Model Security assesses open-weight and proprietary AI models for vulnerabilities and backdoors. It combines model-file scanning, behavioral security testing, compliance mapping and reporting, and side-by-side model benchmarking to help teams identify risks before deployment and document their security posture.

- **Model file security scanning** — Detect Model File Threats
- **Behavioral security testing** — Test Model Resilience
- **Compliance mapping and reporting** — Document Security Compliance
- **Model security benchmarking** — Compare Model Security
- **Remote model source scanning** — Scan Remote Model Sources
- **Broad model format coverage** — Scan Diverse Model Formats
- **AI pipeline integration** — Integrate Security Into Pipelines
- **Security posture tracking** — Track Security Improvements

### [Guardrails](https://www.promptfoo.dev/guardrails/)
*Product*
Protect AI applications from jailbreaks and adversarial attacks with defenses that improve over time and can be independently validated.
Promptfoo Guardrails provides real-time protection against jailbreaks and adversarial attacks. Its adaptive defense system uses red-team findings and actual attack data to improve over time. Teams can use Promptfoo's red-teaming capabilities to independently validate and improve third-party guardrail solutions. Guardrails can be deployed in the cloud or on-premises, with minimal code changes and support for major LLM providers and custom models.

- **Adaptive, self-improving protection** — Strengthen Defenses As Threats Evolve
- **Jailbreak and adversarial attack protection** — Block Jailbreaks And Adversarial Attacks
- **Third-party guardrail validation** — Validate Existing Guardrail Systems
- **Flexible deployment** — Deploy Across Models And Environments

### [MCP Proxy](https://www.promptfoo.dev/mcp/)
*Product*
Control MCP server access, monitor activity, and manage security policies centrally to protect AI applications and users.
MCP Proxy sits between users, AI applications, and MCP servers to provide enterprise-grade security, monitoring, and access control. It lets organizations whitelist approved servers, grant access to specific applications and users, monitor requests with real-time alerts, and manage security policies and audit logs from a centralized dashboard.

- **MCP activity monitoring and alerts** — Track Requests And Detect Suspicious Activity
- **Granular access control** — Limit Access To Required Tools And Data
- **MCP server allowlisting** — Allow Access To Approved MCP Servers
- **Centralized security management** — Manage Policies And Logs In One Dashboard

## Sources

- [Promptfoo](https://promptfoo.dev)
- [Promptfoo Evaluations](https://www.promptfoo.dev/docs/getting-started/)
- [Red Teaming](https://www.promptfoo.dev/red-teaming/)
- [Code Scanning](https://www.promptfoo.dev/code-scanning/)
- [Model Security](https://www.promptfoo.dev/model-security/)
- [Guardrails](https://www.promptfoo.dev/guardrails/)
- [MCP Proxy](https://www.promptfoo.dev/mcp/)
