CrowdStrike and OpenAI are expanding their partnership to secure Codex agents and bring GPT-5.6 Cyber onto the Falcon platform, enabling safer and faster AI adoption across endpoints, SaaS, cloud, and browser environments.

Falcon Guardian now discovers Codex agents, links agent activity to Falcon telemetry in real time, detects and stops threats before they spread, and enforces permissible actions through policy-driven runtime controls.

Within approved defensive use cases, FAIRR will apply GPT-5.6 Cyber to assess risk, analyze attack paths, and prioritize remediation with expert oversight to support defender decision-making at machine speed.

The Claude Marketplace is highlighted as a hub for Claude-compatible tools, emphasizing seamless integration and shared customer commitments.

The press release was published on September 2, 2026, at 1:32 PM EDT by Business Wire.

Charlotte AI is listed in the Claude Marketplace, meeting customers inside the AI environment they already use rather than steering them to a separate portal.

Falcon Model Context Protocol enables Claude and other AI tools to query Falcon data directly for investigations, vulnerability assessments, and operational tasks without switching dashboards.

A Falcon Foundry skills plugin is added to Anthropic’s plugin marketplace, allowing teams to build Falcon applications and workflows using natural language prompts instead of traditional coding.

The collaboration builds on earlier work from April 2026, when Claude Opus 4.7 was used to enhance Falcon’s exposure management and vulnerability prioritization.

Users can deploy Falcon-powered agents from Claude to perform triage, enrichment, hunting, and response, with results returned inline within Claude.

A new Compliance API feeds Claude activity logs and conversations into CrowdStrike’s Next-Gen SIEM and Charlotte SOAR for threat detection and automated responses across endpoints, cloud workloads, and identity systems.

Daybreak features two tracks—Daybreak Blue for defensive workflows and Daybreak Red for offensive security use cases—sharing the same model under enterprise governance.