Bring Cyver Core Directly Into Your AI Workflow
AI assistants are becoming an increasingly useful part of the pentester toolkit: from analyzing technical information and working with code to helping organize findings and automate repetitive tasks.
With the new Cyver Core Model Context Protocol (MCP) Server, pentesters can connect compatible AI assistants directly to Cyver Core and work with pentest data using natural language.
Instead of constantly switching between your AI tools and the pentest management platform, your assistant can securely access permitted Cyver Core data and actions directly within your existing workflow.
The result is a more connected way to use AI throughout the pentesting process—not only for generating content, but for interacting with projects, findings, evidence, assets, reports, and other operational data.
What Is the Cyver Core MCP Server?
The Model Context Protocol (MCP) is an open protocol that enables AI applications to connect to external tools and data sources.
Cyver Core now provides its own MCP Server, giving compatible AI assistants a standardized interface for interacting with the platform.
Once enabled, organizations receive a dedicated MCP Server URL and API key that can be connected to supported AI tools. The actions available to an assistant depend on the tools enabled and the permissions associated with the API key.
Cyver Core currently supports connections with tools including ChatGPT and Claude, opening up new possibilities for AI-assisted pentesting workflows.
More Than Just Asking Questions About Your Data
The MCP Server isn't limited to retrieving information.
Depending on the configured permissions, AI assistants can interact with many of the core objects and workflows inside Cyver Core.
Pentesters can use natural-language instructions to:
- Search and retrieve pentest projects
- Inspect project details, assets, users, teams, and findings
- Create projects from existing templates
- Search and filter findings by severity, status, asset, or vulnerability type
- Retrieve complete findings and supporting evidence
- Create and update findings and evidence
- Summarize vulnerabilities by severity or remediation status
- Work with continuous pentesting projects and runs
- Retrieve report versions and published reports
- Manage clients and assets
- Access project, report, checklist, and compliance templates
- Upload files and import findings from supported vulnerability scanners
This turns an AI assistant into another interface for working with Cyver Core, one built around conversation and intent rather than menus and forms.
Built for the Way Pentesters Work
The biggest benefit of MCP isn't simply connecting another AI tool. It's reducing friction between technical testing, analysis, documentation, and pentest management.
A pentester working inside an AI coding agent can retrieve the context needed from Cyver Core, analyze it, and, where permitted, send structured information back to the platform.
Analyze Findings Without Constant Context Switching
Instead of opening a project, navigating through findings, copying information, and pasting it into an AI assistant, pentesters can request the information directly.
For example:
"List all Critical and High findings in this project."
The assistant can retrieve the relevant information from Cyver Core and use it as context for further analysis.
Work Directly with Finding Evidence
Pentesters can retrieve a finding together with its supporting evidence, analyze the information with their AI assistant, and then update the finding or add new evidence.
This creates a much tighter workflow between technical analysis and documentation.
Turn Technical Work Into Structured Findings
During an assessment, an AI assistant can help transform technical observations into structured pentest data.
A tester could analyze an issue and then instruct the assistant to create a finding in the relevant Cyver Core project, reducing the manual work required to transfer information between testing and reporting tools.
Import Scanner Results Through Your AI Workflow
Cyver Core's MCP tools can also work with supported scanner exports.
Pentesters can upload and import results from tools including Burp Suite, Nessus, Nmap, Qualys, Invicti, OpenVAS, OWASP ZAP, Acunetix, Snyk, Checkmarx, Prowler, Nexpose, and other supported formats.
This makes it possible to incorporate scanner results into a broader AI-assisted workflow while keeping Cyver Core as the central system for managing the resulting findings.
Get Faster Project Context
Before starting or continuing an assessment, a pentester can ask questions such as:
"Summarize the client, status, dates, assets, and finding count for this project."
The assistant can retrieve that information directly, giving the tester fast context without manually navigating multiple areas of the platform.
Practical MCP Use Cases for Pentesters
Connecting an AI assistant to Cyver Core creates opportunities across the full pentest lifecycle.
Project preparation
Retrieve scope, assets, assigned team members, project dates, templates, checklists, and other project information before testing begins.
Testing and analysis
Query existing findings, inspect evidence, identify high-severity issues, and use project information as context while analyzing vulnerabilities.
Finding management
Create new findings, update existing ones, add evidence, and retrieve related technical information without interrupting the testing workflow.
Scanner result processing
Import supported scanner exports and use AI to help review and work with the resulting findings.
Remediation and retesting
Find vulnerabilities waiting for remediation, retrieve their evidence, and update findings as issues are verified or fixed.
Continuous pentesting
Inspect active continuous projects, review run history, retrieve reports, and manage the current testing run.
Reporting and review
Retrieve report versions, summarize findings by severity or status, and quickly access the project information needed during reporting and quality assurance.
AI That Works With Your Pentest Platform
Traditional AI workflows often require pentesters to manually provide context: copying findings, explaining project structures, uploading files, and transferring generated information back into their pentest management platform.
MCP changes that model.
By giving compatible AI assistants controlled access to Cyver Core, the platform itself becomes part of the AI workflow.
Pentesters can spend less time moving information between systems and more time investigating vulnerabilities, validating impact, and delivering useful security insights to clients.
And because access is controlled through the MCP connection and associated API key permissions, organizations remain in control of which Cyver Core capabilities are available to connected assistants.
Connect Your Preferred AI Assistant
The Cyver Core MCP Server can be enabled under Administration > Integrations > MCP Server.
Once enabled, Cyver Core provides the MCP Server URL and API credentials required to connect a compatible assistant.
Connections are currently documented for ChatGPT, Claude Web, and Claude Desktop, making it straightforward to bring Cyver Core data and workflows into the AI environment your pentesters already use.
See the Cyver Core MCP Server in Action
The new MCP Server takes Cyver Core beyond AI-assisted reporting and enables a broader model for AI-assisted pentesting operations.
From retrieving project context and analyzing findings to managing evidence, importing scanner results, and updating pentest data, teams can connect their AI workflows directly with the platform where their engagements are managed.
Ready to connect your AI tools to your pentesting workflow?
Request a demo and discover how the Cyver Core MCP Server can help your pentesters work faster with AI while keeping pentest data and workflows centralized in Cyver Core.
