Keeper Security has launched Model Context Protocol (MCP) integration for its Keeper Secrets Manager, enabling secure and policy-driven interaction between AI agents and sensitive infrastructure secrets. The new feature allows enterprises to adopt AI automation while maintaining zero-trust, zero-knowledge security standards.
SINGAPORE, 23 JULY 2025 – Keeper Security, a global leader in zero-trust and zero-knowledge privileged access management (PAM), has announced the launch of Model Context Protocol (MCP) AI Agent Integration for its Keeper Secrets Manager platform. This new feature allows organizations to securely connect third-party AI agents to manage infrastructure secrets, all while maintaining strict access control and compliance.
As the use of AI assistants grows across enterprise environments, businesses face the challenge of integrating these tools with sensitive systems without compromising security. Keeper’s MCP addresses this by providing a secure, auditable bridge that allows AI tools—whether local or cloud-based—to interact with the Keeper Vault under tightly governed, zero-trust conditions.
The integration supports real-world enterprise demands for automation by enabling AI agents to perform tasks such as generating passwords, managing secrets, attaching files, and conducting system health checks. However, access is governed by least-privilege models and human-in-the-loop approvals to prevent unauthorized activity.
Key features and benefits include:
- Zero-Trust Architecture: Access is limited to explicitly authorized folders and credentials.
- Human-in-the-Loop Oversight: Users must approve sensitive actions in real time.
- Efficiency Boost: Eliminates the need for manual copy-paste of data, reducing errors and security risks.
- Comprehensive Auditing: All interactions with AI agents are logged for full visibility and compliance.
- Cross-Platform Compatibility: Supports Linux, macOS, Windows, and Docker environments.
- Admin-Controlled Enablement: Disabled by default and activated only by administrators via the Keeper Admin Console.
The new integration aligns with Keeper’s broader KeeperPAM® platform, which includes tools to eliminate hard-coded credentials, automate password rotation, and protect digital infrastructure. Keeper Secrets Manager is SOC 2, SOC 3, ISO 27001, 27017, and 27018 certified, FIPS 140-3 validated, and FedRAMP and GovRAMP authorized.
According to Craig Lurey, CTO and Co-founder of Keeper Security, the integration represents a responsible and secure path forward for AI adoption. “AI agents are becoming powerful tools for operational efficiency, but their access to sensitive data must be governed by strong controls,” he said.
Jeremy London, Director of Engineering, AI and Threat Analytics at Keeper, emphasized that MCP provides the secure structure enterprises need to leverage AI without sacrificing compliance or visibility. “We’re giving organizations the ability to automate workflows while maintaining our zero-trust security model and full audit capabilities,” he said.
The launch of MCP integration reinforces Keeper Security’s mission to empower secure innovation across modern enterprises while upholding the highest standards in cybersecurity.
