Bedrock Security Announces Model Context Protocol (MCP) Server to Enable Secure, Context- and Risk-Aware Agentic AI Workflows

Standardizes access to Bedrock Security’s comprehensive metadata lake, accelerating data investigations and enabling smarter data decisions across security, governance and AI systems

MENLO PARK, Calif.--(BUSINESS WIRE)--Bedrock Security, the ubiquitous data security and management platform, announced its Model Context Protocol (MCP) Server at RSAC™ Conference. The MCP Server, to be released during Q2 2025, will enable a secure, standardized gateway between AI agents and enterprise data, auditing model interactions and allowing for safe adoption of open agentic AI standards. It seamlessly integrates deep contextual knowledge of data, risk and usage from the Bedrock Platform’s comprehensive metadata lake directly into enterprise workflows and emerging agentic AI systems, slashing investigation time from weeks to minutes while democratizing deep data insights across the organization. The MCP Server builds upon the Bedrock Metadata Lake Copilot, which makes it simple for teams to interact with the metadata lake and answer complex questions about their data, from identifying sensitive assets to pinpointing riskiest users.

“As organizations adopt agentic AI, they need reliable context to guide AI interactions with enterprise data,” said Pranava Adduri, CTO and co-founder at Bedrock Security. “Our new MCP Server provides standardized access to Bedrock’s metadata lake, delivering detailed information on data sensitivity, risk and usage patterns. This context helps ensure that actions taken by AI agents or within automated workflows align with established policies. It allows teams to integrate AI capabilities more safely and provide valuable feedback as we refine the solution.”

Solving data context and risk fragmentation for the AI era

Enterprises today face pervasive data context fragmentation where important information about data sensitivity, usage, access and risk resides in disconnected silos. This fragmentation fuels operational inefficiency, forcing slow, manual correlation efforts for essential tasks like compliance validation, risk assessment or secure data handling, often bottlenecked by specialized engineering teams.

Bedrock Security’s MCP Server provides a unified, queryable context layer via a standard protocol, making comprehensive data intelligence instantly accessible through simple, iterative queries. Using the open MCP standard, applications and agents can now easily query this rich, interconnected context—over 50 distinct signals—embedding crucial data awareness precisely where needed without disrupting established processes or requiring custom point-to-point integrations.

By overcoming the lack of standardized access to data intelligence, the MCP Server allows broader teams, such as legal and compliance, to make faster, more accurate risk management and governance decisions across complex enterprise environments.

Bedrock Security drives responsible AI implementation through metadata context

With Bedrock Security, organizations can improve security and governance while accelerating innovation by connecting essential context from the metadata lake with AI workflows. For example, an organization could implement an automated sensitive data decommissioning workflow that:

1. Identifies sensitive data in a data warehouse and queries sample records for verification
2. Determines data ownership and identifies users with regular access patterns
3. Automatically notifies relevant stakeholders via Slack to explain why sensitive data is required for their work or whether masked or synthetic variants of the data may suffice
4. Proceeds with automatic decommissioning after predefined periods of inactivity
5. Escalates to human operators when stakeholder input requires further evaluation

“Standardized interfaces like MCP are important for interoperability as data moves between AI agents and tools,” added Adduri. “This isn't just about technical compatibility; it's about practical integration. By supplying specific data context, such as sensitivity, lineage and access details, directly into these workflows via MCP, Bedrock Security helps organizations build systems where security and governance controls can be applied effectively based on reliable data understanding. This supports integrating AI while managing associated risks.”

Bedrock Security continues to deliver capabilities to help enterprises manage the shift to agent-based AI workflows, ensuring governance, traceability and security are embedded by design.

Resources

• Learn more about Bedrock Security
• Request a demo
• Visit our blog
• Follow us on LinkedIn, X and Bluesky

About Bedrock Security

Bedrock Security, the ubiquitous data security and management company, accelerates enterprises’ ability to harness data as a strategic asset while minimizing risk. Its industry-first metadata lake technology and AI-driven automation enable continuous visibility into data location, sensitivity, access and usage across distributed environments. Bedrock’s platform continuously catalogs data, enabling security, governance and data teams to proactively identify risks, enforce policies and optimize data usage — without disrupting operations or driving up costs. Trusted by leading financial institutions, healthcare providers and Fortune 1000 companies, Bedrock Security empowers organizations to improve data security posture management (DSPM), confidently deliver responsible AI initiatives and manage exponential data growth. Headquartered in Silicon Valley and backed by Greylock, the company is led by experts in cloud, GenAI cybersecurity and data storage. Learn more at www.bedrock.security.