Detectify launches MCP Server to secure AI coding loop

Detectify announced the Detectify MCP (Model Context Protocol) Server in a Business Wire press release, describing it as an integration layer that brings Detectify's security testing engines directly into AI-driven development workflows. The company's materials, reported by Business Wire, Help Net Security, SiliconANGLE, and others, list three headline capabilities: "Find & Fix" automation (structured remediation tasks, patch generation, and validation scans), conversational command (natural-language queries of scan results and asset status), and a frictionless, remotely hosted setup for connecting preferred agent tooling.

SiliconANGLE reports the server is built on the Model Context Protocol, the open standard released by Anthropic, and exposes Detectify's scanning engines via MCP so external agents can call scans and retrieve structured findings. Business Wire and SiliconANGLE describe the typical loop as: an agent receives a remediation task, generates a patch, triggers a Detectify validation scan, and then presents the verified result for human review. The vendor materials emphasize a lightweight configuration for connecting the remote MCP Server to third-party agent frameworks.

For practitioners: integrating deterministic security tooling with agent workflows addresses a common gap between probabilistic code generation by LLM-based agents and the need for repeatable vulnerability verification. Industry observers have noted (SiliconANGLE coverage) that MCP is becoming a common interface for agents to reach external services, which lowers integration friction for security vendors that support the standard.

AI-assisted coding is increasing deployment velocity and the volume of code, APIs, and infrastructure that AppSec teams must monitor, a concern highlighted across the press release and reporting. Providing agents with standardized access to existing security scanners can reduce the latency between code change and vulnerability validation, which is important as continuous delivery practices and agentic tooling proliferate. Detectify's prior work on automated testing and asset discovery, documented in Help Net Security and the company blog, indicates this launch is an extension of its continuous-testing focus rather than a wholly new capability set.

Observers will look for how widely agent frameworks and vendor ecosystems adopt MCP for security integrations and whether other AppSec vendors offer equivalent agent-facing endpoints. Analysts and security teams may also monitor operational questions reported by coverage: how validation scans scale when called by many agents, how organizations control agent privileges to invoke remediation workflows, and how the remote-hosted model fits with on-prem or air-gapped environments. Detectify has not provided independent third-party benchmarks of throughput or false-positive rates in the materials cited.

Business Wire and SiliconANGLE both published a direct quote from Detectify CEO Rickard Carlsson: "We aren't competing with the AI's reasoning; we are providing the professional-grade tools that reasoning requires," describing the product as modular building blocks agents can call like test runners.

All coverage reviewed is based on Detectify's announcement or press distribution, and independent technical evaluations or customer case studies were not included in the sources. Reporting also does not quantify scalability, pricing, or enterprise deployment details.