LAS VEGAS, July 29, 2026 (GLOBE NEWSWIRE) — Black Hat USA — Backslash Security, the agentic AI endpoint security company, today announced the Backslash Agentic Fabric Graph, an interactive visualization and risk assessment capability that enables security teams to quickly understand and prioritize risk across the agentic AI fabric – the agents, skills, MCP servers, connectors, and other components – running on employee workstations and laptops. Backslash is also offering a zero-commitment agentless risk assessment, based on the new graph, to qualified organizations who want to quickly understand and prioritize their agentic AI risk.
“Agentic AI usage is skyrocketing, creating a new interconnected ecosystem of AI agents, MCP servers, models, skills, hooks, prompt rules, and connectors on developer and employee desktops – the agentic fabric,” said Shahar Man, co-founder and CEO of Backslash Security. “This has created new, often hidden paths to credentials, internal systems, code, and business data. The Backslash Agentic Fabric Graph brings full visibility to this new and growing attack surface that is largely unseen by security teams.”
The lack of visibility into agentic AI deployments has become a primary hurdle for enterprise security, and a necessary first step for any security program. According to the recent 2026 SANS AI Survey, “sixty-three percent of respondents cite a lack of visibility into where AI models are used and what they expose, up from 56% in 2025 … Without visibility into what AI is doing and without shared standards to judge it, governance stays descriptive instead of prescriptive.”
The Backslash Agentic Fabric Graph provides security teams with a visual, filterable map of the agentic AI endpoint ecosystem. Instead of relying on disconnected inventory lists, security teams can now instantly see how components are connected and where policy violations could have the greatest impact. Using the graph, security teams can answer critical questions in minutes, including:
- Which AI agents are deployed across user workstations, desktops, and laptops?
- Which users have OpenClaw installed, and which MCP servers are they connected to?
- Which MCP server is most widely used in the organization, and does it expose the local network?
- Are there hooks being used that enable data exfiltration?
- What is the potential blast radius of a Skill that runs Bash commands?
- Which users or teams are violating AI security policy or using private accounts on company laptops?
Key features of the Agentic Fabric Graph include:
- Interactive AI Inventory Visualization: Maps AI agents, MCP servers, models, Skills, hooks, and workstations, showing their connections.
- Risk and Policy Violation Filtering: Highlights critical issues, including unapproved models, excessive MCP permissions, shadow AI accounts, and unsafe exposure of local networks.
- User and Workstation-Level Visibility: Identifies private account use, policy violations, and agent sprawl at the user and workstation levels.
- Risk Prioritization: Helps teams determine which components are most widely used and where remediation will have the highest impact.
The Graph is a visual layer built into the broader Backslash Agentic AI Endpoint Security platform. The full platform provides comprehensive discovery, governance, and real-time protection across interconnected AI coding, cowork agents, and other fabric components. It utilizes intent-based analysis to inspect agent behavior in real time, infer the intent behind each action, and block harmful activity inline – including prompt injections, data exfiltration, privilege escalations, and agents going rogue. This enables security teams to move beyond chasing risks after the fact to intercepting them as they happen.
Limited-Time Risk Assessment Offer:
To help organizations immediately uncover their current risks, Backslash is offering a no-commitment AI Endpoint Exposure Assessment, designed for enterprises with fleets of 200+ endpoints. This guided, agentless, and read-only assessment maps every agent, MCP, skill, and hook across the organization’s endpoints. Within 24 hours, organizations will receive a board-ready visualization and report detailing their security posture score, critical findings, and a prioritized remediation plan.
To apply, visit: https://www.backslash.security/ai-endpoint-exposure-assessment
Join Backslash at Black Hat USA 2026
Backslash Security will showcase the Agentic Fabric Risk Graph and its full agentic endpoint security platform at Black Hat USA 2026. Attendees can meet the team at:
- Black Hat AI Summit, Aug. 3rd
- Black Hat Business Hall, Aug. 3rd-5th at Booth 4909 in the AI Zone.
To pre-book a meeting with the Backslash executive team, visit https://www.backslash.security/blackhat-2026
About Backslash Security
Backslash is the Agentic AI Endpoint Security platform protecting enterprise developer and employee endpoints. Backslash empowers security and AI governance teams to accelerate the adoption of AI coding and agentic AI at scale, while providing full visibility, governance, and real-time protection across the new attack surface of interconnected AI coding agents, MCP servers, IDEs, plugins, skills, and hooks. Backslash detects, prevents, and responds to threats, including prompt injection, data and source exfiltration, malicious and compromised MCPs, and abuse of AI agent privileges. For more information, visit https://backslash.security.
Media Contact:
Doug De Orchis
Scratch Marketing & Media for Backslash
[email protected]
A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/1afe8ac7-4812-4e1b-86f7-5f6cf94f65b4