promptfoo-dev

Skip to main contentPromptfoo has joined OpenAI. Read the update →Top 5 global retailerTop 3 U.S. wireless carrierTop 3 global strategy consultancyPublic legal & tax providerTop 5 U.S. healthcare companyNationwide 5G network operatorTop foundation model labTop 3 U.K. insurance providerGlobal ERP leaderTop 3 travel siteTop 5 global retailerTop 3 U.S. wireless carrierTop 3 global strategy consultancyPublic legal & tax providerTop 5 U.S. healthcare companyNationwide 5G network operatorTop foundation model labTop 3 U.K. insurance providerGlobal ERP leaderTop 3 travel site127 of the Fortune 500 use Promptfoo in their AI development lifecycle. Here's how:Red TeamingEvaluationsCode ScanningModel SecurityGuardrailsMCPAutomated red teaming for agents & RAGsPromptfoo simulates real users to uncover application-specific vulnerabilities:npx promptfoo@latest redteam setupCustom attacks generated for your target:Direct and indirect prompt injectionsJailbreaks tailored to your guardrailsData and PII leaksBusiness rule violationsInsecure tool use in agentsToxic content generationAnd much more» Learn more about Red TeamingTHE PROMPTFOO APPROACHAI security testing built intoyour development workflowFrom integration to remediation, Promptfoo meets you wherever you're building.1ConnectIntegrate AnywhereConnect to your AI apps, agents, and workflows.CI/CD pipelinesGitHub, GitLab, Jenkins, and moreMCP and Agent frameworksOn-premise or cloudLearn about integrations →2AttackTest EverythingCreate thousands of context-aware attacks tailored to your application.Real-time threat intel from 300k+ user communityDeep automation that scales beyond human-curated testsCustomize attack flows to your businessExplore red teaming →3FixClose the LoopGet remediation guidance directly in pull requests and developer workflows.Security findings in PRsActionable remediation stepsTrack fixes across teamsContinuous monitoringLearn more about remediation →THREAT INTELLIGENCE AT GLOBAL SCALEBuilt on the world's largestAI security communityGet real-time threat intelligence and innovation you can't find anywhere else.Our contributors are from companies like OpenAI, Google, Microsoft, and Amazon.0+Open Source UsersDevelopers securing AI applications with Promptfoo0+ContributorsFrom major foundation labs and tech companies0+Weekly DownloadsActive deployments in production workflows worldwideUsed by the bestTrusted by AI LeadersSee how teams at OpenAI and Anthropic use Promptfoo to build better AI applications.Build Hours"Promptfoo is really powerful because you can iterate on prompts, configure tests in YAML, and view everything locally... it's faster and more straightforward"Watch the Video →Courses"Promptfoo offers a streamlined, out-of-the-box solution that can significantly reduce the time and effort required for comprehensive prompt testing."See the Course →SOLUTIONS FOR EVERY TEAMSecurity that works for everyoneWhether you're building AI security strategy or writing code, Promptfoo meets you where you are.For CISOsStrategy & RiskFor Security DirectorsDepth & AutomationFor DevelopersSpeed & EnablementFor Security DirectorsDepth & AutomationYou need a solution that actually works at enterprise scale, integrates with your existing tools, and your team will adopt. Promptfoo delivers the depth you need without the complexity.What You Get:Proven at global scale - 127 of the Fortune 500Leading healthcare, telecommunications, retail, and enterprise software companies trust us with their AI security.Application-focused, not just model testingTests understand your business logic, RAG, agents, integrations. Covers 50+ vulnerability types from injection to jailbreaks.Deep automation that actually scalesNo manual scenario writing required. Continuous testing in CI/CD. Scales from 1 to 100+ applications.Real-time threat intelligenceCommunity of 300k+ users provides early warning. New attack vectors deployed automatically.Schedule DemoShip Secure, Stay SecureJoin hundreds of enterprises and thousands of developers securing AI applications from day one.Request DemoTry Open Source✓ 300,000+ developers • ✓ Enterprise trusted • ✓ Zero vendor lock-in We use cookies for analytics to improve our site. Privacy policyDeclineAccept --- Skip to main contentPromptfoo has joined OpenAI. Read the update →CommunityFree ForeverOur open-source tool, perfect for individual developers and small teams.All LLM evaluation featuresAll model providers and integrationsRed teaming (10k probes/month)Custom integration with your own appRun locally or self-host on your own infrastructureVulnerability scanningCommunity supportGet StartedEnterpriseCustomFor teams that need advanced features. Learn moreAll Community featuresCustom red teaming limitsTeam sharing & collaborationContinuous monitoringCentralized security/compliance dashboardCustomizable attack profiles and target settingsSSO and granular permission profilesPromptfoo API accessManaged cloud deploymentProfessional services supportPriority support & SLA guaranteesSchedule DemoOn-PremiseCustomFor organizations that require full control over their infrastructure.All Enterprise featuresDeployment on your own infrastructureComplete data isolationDedicated runnerAssigned deployment engineerContact UsEnterprise FeaturesReports & Continuous MonitoringUnderstand your LLM security status across all projects with real-time alerts and automated evaluations.Issue Tracking & Guided RemediationTrack remediation progress and get suggested steps for each issue.Comprehensive Scanning & ComplianceVerify compliance with industry frameworks and standards.Organization-Specific ConfigurationsCreate customizable plugin collections, scan configurations, and target settings that can be shared among colleagues.Teams-Based ControlsManage your LLM applications with teams-based access control and SSO, granular permission profiles, and customizable API access.Feature ComparisonFeatureCommunityEnterpriseEnterprise On-PremiseSecurity & TestingLLM evaluation capabilitiesBasic vulnerability scanningAdvanced vulnerability detection—Remediation recommendations—Centralized guardrail dashboard—Customized red teaming plugins—Organization-specific attack profiles—Saved target configurations—Searchable scan history—Team & CollaborationTeams-based access control and SSO—Team sharing & collaboration—Custom roles & permissions—Centralized dashboard—IntegrationsCI/CD integrationPromptfoo API integration—Webhooks—InfrastructureCloud deployment——On-premise deployment——Complete data isolation——Support & ServicesCommunity supportPriority support & SLA guarantees—Dedicated support team—Dedicated deployment engineer——Frequently Asked QuestionsWhat's included in the Community version?The Community version includes all core features for local testing, evaluation, and vulnerability scanning.What is a probe?A probe is a single request made to your target system during red team testing.Why are there usage limits on red teaming?Certain red teaming plugins require inference for dynamic test generation and grading. The open-source Community version includes up to 10k probes per month at no charge. Enterprise customers can purchase additional probes for larger scale testing.Who needs the Enterprise version?Larger teams and organizations that want to continuously monitor risk in development and production.How does Enterprise pricing work?Enterprise pricing is customized based on your team's size and needs. Contact us for a personalized quote.Can I upgrade from Community to Enterprise?Yes, you can easily upgrade to Enterprise at any time to access additional features and support.Ready to get started?Schedule a demo to see how Promptfoo can help secure your LLM applications.Schedule a Demo We use cookies for analytics to improve our site. Privacy policyDeclineAccept --- Skip to main contentPromptfoo has joined OpenAI. Read the update →Securing the Future of AIPromptfoo helps developers and enterprises build secure, reliable AI applications.About UsWe are security and engineering practitioners who have scaled generative AI products to hundreds of millions of users. We're building the tools that we wished we had when we were on the front lines.Based in San Mateo, California, Promptfoo joined OpenAI on March 16, 2026. Promptfoo remains open source, and we continue to build tools for secure, reliable AI applications.Our foundersIan WebsterCEO & Co-founderIan previously led LLM engineering and developer platform teams at Discord, scaling AI products to 200M users while maintaining rigorous safety, security, and policy standards.Michael D'AngeloCTO & Co-founderMichael brings extensive experience in AI and engineering leadership. As the former VP of Engineering and Head of AI at Smile Identity, he has a track record of scaling ML solutions to serve over 100 million people across hundreds of enterprises.Early SupportersWe're grateful to the investors and operators who backed Promptfoo early and helped us build open-source, application-focused AI security.Ganesh BellManaging Director, Insight PartnersZane LackeyGeneral Partner, Andreessen Horowitz Founder, Signal SciencesJoel de la GarzaInvestment Partner, Andreessen Horowitz CISO, BoxTobi LutkeCEO, ShopifyStanislav VishnevskiyCTO, DiscordFrederic KerrestVice-Chairman & Co-Founder, OktaAdam ElyEVP, Head of Digital Products, Fidelity CISO, FidelityAn Incredible Open Source CommunityPromptfoo is proud to be supported by a vibrant community of over 261 open source contributors.Ready to Secure Your AI Applications?Join leading enterprises who trust Promptfoo to fortify their AI applications.Get in Touch We use cookies for analytics to improve our site. Privacy policyDeclineAccept --- Skip to main contentPromptfoo has joined OpenAI. Read the update →Today we are announcing that Promptfoo has agreed to be acquired by OpenAI. Promptfoo will remain open source and we will continue to serve users and customers. We founded Promptfoo in 2024 to make it easy for developers to systematically test their AI applications. We quickly realized that adversarial tests for security, safety, and other behavioral risks were the biggest blockers to shipping AI, especially at large enterprises. What we built grew faster than we ever could have imagined. More than 350k developers have used it, 130k are active each month, and teams at more than 25% of the Fortune 500 rely on it. We are joining OpenAI so that the security, evaluation, and compliance platform we've built - and the frontline experience behind it - can have the greatest impact on how teams build and deploy AI. At OpenAI, we'll improve and integrate Promptfoo's core tech within the model and infrastructure layers, so teams can catch vulnerabilities early and ship secure AI from the start. What this means for Promptfoo users​ OpenAI gives our work more resources and access to research at the model and inference layers that supercharge our goal of helping everyone ship secure, reliable AI. This is the fastest and most impactful path forward for the work we started at Promptfoo. The team will continue working with customers and users to ensure continuity of service and support. We will continue to maintain the open-source suite as a best-in-class red teaming, static scanning, and evals tool for any AI model or application. Promptfoo will continue to support a diverse range of providers and models, reflecting the way real teams build and deploy AI systems. Thank you​ We have so much gratitude toward our investors: Ganesh at Insight Partners, Zane at a16z, their teams, and all the other angel investors that supported us. You helped us scale Promptfoo faster than we ever thought possible. We are grateful for our team: we grew quickly to 23 people across engineering, GTM, and operations - the most talented and hard-working bunch we've ever met. Finally, thank you to everyone who contributed code, filed issues, uses the product, or trusts Promptfoo in production. You helped build something important. We're excited to continue this work. Ian Webster and Michael D'AngeloCo-founders, Promptfoo The closing of the acquisition is subject to customary closing conditions. Promptfoo built what we believe is a category-defining platform for AI evaluation and security. As enterprises deploy more complex AI systems, rigorous testing, red teaming, and evaluation become foundational. Ian, Michael, and the team built something essential.Ganesh BellManaging Director, Insight PartnersWe believed early that AI security would become mission-critical, and Promptfoo validated that thesis in a big way. Ian, Michael, and the team built a platform that helps organizations find and fix AI risks before they ship, all while building in the open and earning deep trust from developers and enterprises alike. We're incredibly excited to see their continued impact on the future of AI security.Zane LackeyGeneral Partner, Andreessen HorowitzWhat this means for Promptfoo usersThank you We use cookies for analytics to improve our site. Privacy policyDeclineAccept