Detectify debuts MCP server to let AI agents find and fix vulnerabilities in real time

Application security platform company Detectify AB today launched the Detectify MCP Server, a brand new integration layer that plugs the corporate’s security testing engines into artificial intelligence-driven coding workflows in order that agents can find, validate and remediate exploitable vulnerabilities in real time.

Detectify’s MCP Server is built on the Model Context Protocol, the open standard Anthropic PBC released in November 2024 that has since been adopted across the AI industry because the default way for agents to consult with outside tools. The corporate is pitching the launch at an issue security teams have flagged for months, namely that AI coding agents now ship code faster than human-led review cycles can sustain with.

By exposing its scanning engines through MCP, Detectify is offering agents a standardized method to query findings and trigger validation as a part of autonomous development loops reasonably than waiting for periodic human review. The corporate argues that AI-assisted coding is concurrently eliminating some common errors while dramatically expanding the quantity of software, application programming interfaces and infrastructure that organizations must track. The issue is compounded by shadow information technology and shadow AI adoption inside enterprises.

The MCP Server ships with two headline capabilities. A “Find & Fix” automation feature hands security findings to AI agents as structured remediation tasks. It allows an agent to generate a patch, trigger a Detectify validation scan to verify the fix and surface the result for human review. A conversational interface lets users query scan results, monitor asset status and surface high-severity findings through natural-language prompts. The server is remotely hosted, with what Detectify describes as a light-weight configuration for connecting to preferred AI tools.

“We aren’t competing with the AI’s reasoning, we’re providing the professional-grade tools that reasoning requires,” said Chief Executive Rickard Carlsson. “By structuring our capabilities as modular, high-performance constructing blocks, we allow agents to call our scanner as naturally as they call a test runner. We’re expanding from a dashboard humans check to a skill agents orchestrate.”

Carlsson framed the broader argument by way of determinism. Large language models excel at reasoning but operate probabilistically, meaning they don’t produce the identical answer to a question every time. Detectify is pitching its compiled scanning engines because the deterministic verification layer agents need before code reaches production. The corporate monitors thousands and thousands of fixing domains and says the MCP Server is meant to bring that scale into agentic workflows so security can operate at engineering velocity.

The launch comes amid a wave of comparable MCP server releases from security vendors over the past 12 months, including Legit Security Ltd., JFrog Ltd. and TrojAI Inc., as application security firms race to make their tooling addressable by autonomous coding agents.

The Detectify MCP Server shall be available soon as a part of Detectify’s continued investment in AI-native application security.

Image: Detectify