Kong has launched Insomnia 12, a major update to its open-source API development platform designed to streamline both API and Model Context Protocol (MCP) server development. With a focus on accelerating testing, iteration, and deployment, the new release incorporates AI-powered collaboration and automation features to reduce manual overhead and help teams deliver more reliable services faster.
Announced on November 4 and now generally available, Insomnia 12 introduces native MCP clients, AI mock servers, and AI-driven commit suggestions. These enhancements aim to simplify validation and testing workflows for MCP servers, eliminating the need for complex setup while maintaining high accuracy. The platform now extends its traditional test-iterate-debug workflow into AI-native development, enabling developers to more efficiently build and deploy MCP services.
Among the key capabilities, native MCP clients allow developers to connect directly to servers, manually invoke tools or prompts, and inspect messages at the protocol and authentication level. AI mock generation makes it possible to create mock servers simply by describing requirements in natural language or providing a URL, JSON sample, or OpenAPI specification. Meanwhile, AI-powered commits automatically generate descriptive commit messages and logical file groupings by analyzing code diffs and history, streamlining version control and collaboration. Teams can choose between cloud-based or local LLMs, providing flexibility to balance performance, privacy, and data residency requirements.
Insomnia 12 also enhances team collaboration and governance. Git Sync enables seamless cross-machine development, while enterprise users can trial advanced features like SCIM, SSO, and RBAC for security, compliance, and access management. By combining AI-driven tools with robust collaboration capabilities, Insomnia 12 positions itself as a powerful platform for teams looking to accelerate both API and MCP server development while maintaining control, visibility, and reliability across projects.