CISO Tribune

Vendor comparison · 2 companies

Onyx Security vs Promptfoo

Onyx Security sells a secure control plane for AI agents and models. Promptfoo builds open-source tools for secure, reliable AI applications. CISO Tribune files both under AI model and agent security. This page sets the two records side by side: what each sells, where it is based, who owns and funds it, the customers it names and its recent news, each fact numbered to its source.

Change the companies

Up to four companies side by side.

  • Onyx Security
  • Promptfoo
Onyx Security vs Promptfoo: Products, Customers and Funding Compared
Onyx SecurityOnyx Security sells a secure control plane for AI agents and models.onyx.securityPromptfooPromptfoo builds open-source tools for secure, reliable AI applications.promptfoo.dev
What it sells
  • AI Security[1][7]Onyx provides the runtime and posture layer of the Onyx Secure AI Control Plane.
  • MCP Gateway[8]The MCP Gateway governs the agent-to-tool surface.
  • Red Teaming[2][4][5]Simulates real users to uncover application-specific vulnerabilities.
  • Evaluations[2][5]LLM evaluation of prompts and models that can run inside CI workflows.
Categories
Headquarters
Not stated on its pagesSan Francisco, United States[1][3]
Founded
Not stated on its pages2024[1][12]
Ownership and funding
Not stated on its pagesOwned by OpenAI[12]
Named customers
Not stated on its pagesNot stated on its pages
Recent news
  • July 29, 2026Onyx Security announced a $113 million Series B funding round led by Bessemer Venture Partners.[15]
  • March 12, 2026Onyx Security officially launched operations after securing $40 million in initial funding.[14]
  • March 9, 2026OpenAI agreed to buy Promptfoo, a startup helping businesses find and fix security issues in AI models during development.[13]
  • March 9, 2026OpenAI announced it has acquired Promptfoo, an AI security startup founded in 2024 to protect LLMs from adversarial attacks.[12]
G2 and Gartner Peer Insights
  • G2: Not stated on its pages
  • Gartner Peer Insights: Not stated on its pages
  • G2: Not stated on its pages
  • Gartner Peer Insights: Not stated on its pages
Who it fits
From its own pages, Onyx Security sells mainly to security, governance, and AI teams that need control over AI agent adoption across the enterprise. Onyx Security describes a single platform for security, governance, and infrastructure teams. Onyx Security offers cloud, hybrid, or self-hosted options for organisations with sensitive data residency requirements. Onyx Security presents one console for AI agents and models across SaaS, cloud, endpoint, and code environments. Onyx Security names security, finance, and platform teams as the teams that share its usage, cost, and policy ledger.Not written yet
Last verified
October 19, 2026November 4, 2026

Where they overlap

  • AI model and agent security is the primary category of both Onyx Security and Promptfoo.

Where they differ, field by field

  • Onyx Security is also filed under AI governance, Workforce AI usage security and Non-human identity security. Promptfoo is also filed under AI firewalls and guardrails, Static application security testing and Breach simulation and automated pen-testing.
  • Onyx Security lists AI Security and MCP Gateway. Promptfoo lists Red Teaming and Evaluations.
  • Promptfoo is headquartered in San Francisco, United States.
  • Promptfoo was founded in 2024.
  • Promptfoo is owned by OpenAI.

How they differ

CISO Tribune files both Onyx Security and Promptfoo under AI model and agent security as the primary category. Onyx Security sells a secure control plane for AI agents and models, including an AI Security product that inspects prompts, tool calls, and model responses inline, and an MCP Gateway that governs the agent-to-tool surface. Promptfoo sells Red Teaming, which simulates real users to uncover application-specific vulnerabilities, and Evaluations, an LLM evaluation of prompts and models that can run inside CI workflows. Both companies touch red teaming of agents: Onyx says its red teaming continuously attacks deployed agents, while Promptfoo creates thousands of context-aware attacks tailored to the customer's application.

Onyx Security describes a single platform for security, governance, and infrastructure teams, with cloud, hybrid, or self-hosted options. Promptfoo builds open-source tools, and its Community version includes up to 10k probes per month at no charge. Promptfoo's evaluations can be integrated into a CircleCI pipeline and with test frameworks like Jest and Vitest. Onyx Security names Okta, Microsoft Entra, Splunk, and CrowdStrike NG-SIEM among its integrations.

Promptfoo is headquartered in San Francisco, United States, was founded in 2024, and is owned by OpenAI. OpenAI announced on March 9, 2026 that it had acquired Promptfoo. Onyx Security officially launched operations on March 12, 2026, after securing $40 million in initial funding. On July 29, 2026, Onyx Security announced a $113 million Series B funding round led by Bessemer Venture Partners.

What it sells: Onyx Security

What AI Security does[1][7]
Onyx provides the runtime and posture layer of the Onyx Secure AI Control Plane. Onyx inspects every prompt, tool call, and model response inline across browser AI, coding assistants, desktop AI agents, cloud workloads, and MCP servers. Onyx offers five enforcement actions: alert, block, mask, steer, or ask.
What MCP Gateway does[8]
The MCP Gateway governs the agent-to-tool surface. Every tool call an agent makes to an MCP server gets inspected and policy-enforced. MCP Gateway coverage spans the MCP ecosystem, including servers using dynamic client registration and popular servers like GitHub.

What it sells: Promptfoo

What Red Teaming does[2][4][5]
Promptfoo creates thousands of context-aware attacks tailored to the customer's application. The open-source Community version includes up to 10k probes per month at no charge.
What Evaluations does[2][5]
Promptfoo's LLM evaluation can be integrated into a CircleCI pipeline. Promptfoo can be integrated with test frameworks like Jest and Vitest to evaluate prompts as part of existing testing and CI workflows. Promptfoo provides real-time alerts and automated evaluations to show LLM security status across all projects.

Questions about Onyx Security and Promptfoo

What do Onyx Security and Promptfoo sell?
Onyx Security sells AI Security and MCP Gateway. Promptfoo sells Red Teaming and Evaluations.
Are Onyx Security and Promptfoo in the same category?
Yes. AI model and agent security is the primary category of both on CISO Tribune.
Where are Onyx Security and Promptfoo headquartered?
Promptfoo is headquartered in San Francisco, United States.
When were Onyx Security and Promptfoo founded?
Promptfoo was founded in 2024.
Who owns Onyx Security and Promptfoo?
Promptfoo is owned by OpenAI.
Who are Onyx Security and Promptfoo sold to?
From its own pages, Onyx Security sells mainly to security, governance, and AI teams that need control over AI agent adoption across the enterprise.
What have Onyx Security and Promptfoo announced recently?
July 29, 2026: Onyx Security announced a $113 million Series B funding round led by Bessemer Venture Partners. March 9, 2026: OpenAI agreed to buy Promptfoo, a startup helping businesses find and fix security issues in AI models during development.

About Onyx Security

Onyx Security sells a secure control plane for AI agents and models. Onyx gives security, governance, and AI teams control to protect, optimize, and measure AI agent adoption across the enterprise. Onyx was founded around a single goal: to control the most advanced forms of AI. Onyx describes its secure AI control plane as a single platform that serves security, governance, and infrastructure teams as they oversee how AI agents are used, ensure they are secure and compliant, and measure their effectiveness.[1][2]

Onyx says enterprises were deploying autonomous agents faster than they could see, govern, or control them. Onyx offers one console for every AI agent and model across SaaS, cloud, endpoint, and code. Onyx discovers sanctioned and shadow AI across the enterprise, with visibility into every prompt, response, and agent action. Onyx identifies every agent, MCP server, AI tool, copilot, and embedded AI surface continuously, across SaaS, cloud, endpoints, and code. Onyx combines directory scanning, native platform APIs, and behavioral signature learning for browser-based AI to keep the inventory current. Onyx records every prompt, model response, and tool call by name, argument, and result, with the reasoning context preserved across turns. Onyx correlates agent activity to the invoking user and the agent owner, and captures the identity used on each tool and MCP call. Onyx forwards alerts and their session records to Splunk and CrowdStrike NG-SIEM.[1][2][4][5][8]

Onyx provides the runtime and posture layer of the Onyx Secure AI Control Plane. Onyx inspects every prompt, tool call, and model response inline across browser AI, coding assistants, desktop AI agents, cloud workloads, and MCP servers. Onyx applies five enforcement actions: alert, block, mask, steer, or ask (human in-the-loop). Onyx evaluates every agent on deployment and continuously across its lifecycle. Onyx red teaming continuously attacks deployed agents, generating attack plans across reconnaissance, jailbreaking, and weaponization. Onyx identifies supply chain risks in agents, MCP servers, models, and AI assets.[4][6]

Onyx gives an agent its own identity, separate from the invoker, when multiple users delegate to the same agent across tools. Onyx lets teams sanction approved tools and control MCP server access. Onyx offers pre-populated policy templates aligned to industry regulations. Onyx helps teams meet security standards from MITRE, NIST, OWASP, and more.[4][7]

Onyx centralizes all AI traffic on a fully-managed, OSS AI gateway. Onyx routes across popular LLMs to automate failover, minimize latency, and balance load by task complexity and cost constraints. Onyx proxies all MCP traffic with an inline gateway, logs every request and response, and enforces guardrails. Onyx evaluates every tool call an agent makes to an MCP server against scope, identity, data class, and policy before the call lands.[4][8]

Onyx lets teams define AI adoption targets by department, use case, or tool category. Onyx shows total AI spend, avoidable costs, and optimization opportunities. Onyx correlates AI usage, cost, risk, and outcomes to the same activity record. The Onyx AI Gateway captures per-request cost metadata at runtime: cost, tokens, model, and identity context. The Onyx Guardian Agent is a supervisory AI that continuously works across the platform to identify risks and remediate issues.[2][4][9]

Onyx offers cloud, hybrid, or self-hosted options for sensitive data residency requirements. Onyx supports deployment with AWS VPC, Bedrock Gateway, and custom proxy configurations. Onyx offers 100+ pre-built integrations that connect to agent platforms, frontier model providers, SIEM, and cloud platforms. Onyx pulls identity context from Okta and Microsoft Entra, and endpoint management context from Intune, Jamf, and Tanium. Onyx integrates with Claude Enterprise through the Compliance API, pulling conversation content, uploaded files, project data, and activity logs into the Onyx control plane. Onyx integrates with the OpenAI Compliance API, ingesting ChatGPT Enterprise workspace activity. Onyx integrates with Amazon Bedrock Agents through Bedrock Agent configuration APIs, agent invocation logs delivered to S3, and CloudTrail data events. Onyx integrates with Microsoft Copilot Studio through the Microsoft Graph API. Onyx Security is SOC 2 Type II and ISO 27001 certified.[1][4][6][7][8][10][11][12][13]

Onyx Security on the record

Sources

Onyx Security full record

  1. [1]onyx.securityOnyx Security · onyx.security · undated page, accessed October 19, 2026 · Tier A
  2. [2]The Secure AI Control Plane for the Agentic Era | OnyxOnyx Security · undated page, accessed November 10, 2026 · Tier A
  3. [3]About Onyx | The Team Behind the AI Control PlaneOnyx Security · undated page, accessed November 10, 2026 · Tier A
  4. [4]One Platform to Discover and Control Every AI Agent | OnyxOnyx Security · undated page, accessed November 10, 2026 · Tier A
  5. [5]AI Observability – Discover Every Agent & Session | OnyxOnyx Security · undated page, accessed November 10, 2026 · Tier A
  6. [6]AI Agent Security | Block Prompt Injection in Real TimeOnyx Security · undated page, accessed November 10, 2026 · Tier A
  7. [7]AI Governance Platform Built for the EU AI Act | OnyxOnyx Security · undated page, accessed November 10, 2026 · Tier A
  8. [8]AI Orchestration – Governed Routing & Control | OnyxOnyx Security · undated page, accessed November 10, 2026 · Tier A
  9. [9]AI ROI Platform – Prove Value & Control Costs | OnyxOnyx Security · undated page, accessed November 10, 2026 · Tier A
  10. [10]Secure Claude Enterprise with Onyx | Anthropic IntegrationOnyx Security · undated page, accessed November 10, 2026 · Tier A
  11. [11]Secure ChatGPT Enterprise with Onyx | OpenAI IntegrationOnyx Security · undated page, accessed November 10, 2026 · Tier A
  12. [12]Secure Amazon Bedrock Agents | Onyx IntegrationOnyx Security · undated page, accessed November 10, 2026 · Tier A
  13. [13]Copilot Studio Security – Discover, Govern & Control | OnyxOnyx Security · undated page, accessed November 10, 2026 · Tier A
  14. [14]Israeli cyber firm Onyx Security launches operations with $40 million funding round | ReutersReuters · published March 12, 2026 · Tier B
  15. [15]Onyx Security Raises $113M Series B to Control Advanced AI, Quadrupling Revenue since Stealth Launch Four Months AgoBusiness Wire · published July 29, 2026 · Tier A

Promptfoo full record

  1. [1]PressPromptfoo · promptfoo.dev · undated page, accessed November 4, 2026 · Tier A
  2. [2]promptfoo.devPromptfoo · promptfoo.dev · undated page, accessed November 4, 2026 · Tier A
  3. [3]About Promptfoo | AI Security Experts | PromptfooPromptfoo · undated page, accessed November 10, 2026 · Tier A
  4. [4]Build Secure AI Applications | PromptfooPromptfoo · undated page, accessed November 10, 2026 · Tier A
  5. [5]Pricing | PromptfooPromptfoo · undated page, accessed November 10, 2026 · Tier A
  6. [6]Setting up Promptfoo with CircleCI | PromptfooPromptfoo · undated page, accessed November 10, 2026 · Tier A
  7. [7]Testing prompts with Jest and Vitest | PromptfooPromptfoo · undated page, accessed November 10, 2026 · Tier A
  8. [8]Finding LLM Jailbreaks with Burp Suite | PromptfooPromptfoo · undated page, accessed November 10, 2026 · Tier A
  9. [9]SharePoint Integration | PromptfooPromptfoo · undated page, accessed November 10, 2026 · Tier A
  10. [10]Langfuse integration | PromptfooPromptfoo · undated page, accessed November 10, 2026 · Tier A
  11. [11]AI Security for Real Estate | PromptfooPromptfoo · undated page, accessed November 10, 2026 · Tier A
  12. [12]OpenAI acquires Promptfoo to secure its AI agents | TechCrunchTechCrunch · published March 9, 2026 · Tier B
  13. [13]OpenAI Buying AI Security Startup Promptfoo to Safeguard AI Agents - BloombergBloomberg · published March 9, 2026 · Tier B

Spot a factual error? Request a correction.