Find notable cyber news and cases, enriched with sources, timelines, and signals.
Home Browse News Feed Stats Watchlists Beta About Contact

OpenAI integrates Promptfoo agent security testing into Frontier

Security Tool/Service
First reported
Last updated
Happening score
H score 10
1 unique sources, 1 articles

Summary

Hide ▲

OpenAI is adding Promptfoo-style agent security testing into Frontier, turning automated red-teaming and risk checks into built-in capabilities for enterprise AI agents. The change matters because it is meant to catch prompt injections, jailbreaks, data leaks, and tool misuse before deployment. It also pushes security and evaluation earlier in the development workflow, giving organizations more visibility into agent behavior and governance. The integration signals a broader move to make security a native part of enterprise AI operations.

Related Happenings

Google GTIG analysis of adversary AI use for exploit development and attack orchestration

Technical Analysis
First: 11.05.2026 16:00 Last: 11.05.2026 16:00 Sources 1

About this happening: **Google Threat Intelligence Group** published findings showing **adversaries using AI** for **exploit development** and **attack orchestration**, signaling that model-assisted tr...

Open Happening

OpenAI expands Trusted Access for Cyber to government cyber defenders

Security Tool/Service
First: 04.05.2026 11:00 Last: 04.05.2026 11:00 Sources 1

About this happening: **OpenAI** expanded its **Trusted Access for Cyber (TAC)** program to **federal, state, and local governments**, broadening a defender-access system for cyber response and critica...

Open Happening

Gemini Enterprise Agent Platform launch adds agent identity, policy enforcement, and anomaly detection controls

Security Tool/Service
First: 23.04.2026 15:00 Last: 23.04.2026 15:00 Sources 1

About this happening: Google Cloud expanded **Gemini Enterprise Agent Platform** with new security controls for **AI agents**, giving organizations more visibility and policy enforcement for autonomous...

Open Happening

OpenAI launches GPT‑5.4‑Cyber and expands TAC access for cyber defense

Security Tool/Service
First: 15.04.2026 19:00 Last: 15.04.2026 19:00 Sources 1

About this happening: OpenAI launched **GPT‑5.4‑Cyber** and expanded **Trusted Access for Cyber (TAC)**, giving vetted defenders broader access to a **cyber-permissive** model for **defensive workflows...

Open Happening

OpenClaw hardening guidance (CNCERT)

Advisory/Mitigation
First: 14.03.2026 18:17 Last: 14.03.2026 18:17 Sources 1

About this happening: China's **CNCERT** issued mitigation guidance for **OpenClaw**, warning that weak defaults and privileged access could let attackers seize endpoints, leak data, or trigger destruc...

Open Happening

Timeline

  1. 10.03.2026 19:15 1 articles · 2mo ago

    OpenClaw advisor calls for AI-tool scanning against human-language malware

    Technical Analysis Update

    Jamieson O’Reilly, security advisor at OpenClaw and founder of DVULN, said the AI and cybersecurity community needs better ways to scan AI tools for human-language malware rather than relying on traditional file-based malware analysis.

    Show sources
    Open in new tab
  2. 10.03.2026 19:15 2 articles · 2mo ago

    OpenAI acquires Promptfoo and plans Frontier security integration

    Initial Disclosure

    OpenAI said it is acquiring Promptfoo to strengthen agentic AI security testing for OpenAI Frontier, with automated security testing, red-teaming, reporting, and traceability designed to identify prompt injections, jailbreaks, data leaks, tool misuse, and out-of-policy agent behaviors before deployment; OpenAI also said Promptfoo will remain open source.

    Show sources
    Open in new tab