Show HN: StrictLLM – For when you need human-in-the-loop (strictllm.com)

StrictLLM has launched a new platform that allows users to coordinate multiple language models (LLMs) within a single workspace, enhancing the human-in-the-loop approach for AI workflows. Users can directly compare responses from different models, invoke specialized personas through @mentions, and manage tasks across varied AI models, all while maintaining a clear and auditable process. This innovative orchestration enables distinct personas tailored to specific roles—such as coding or writing—ensuring a fluid interaction among AI components. The significance of StrictLLM lies in its enterprise-grade security features, including granular tool-level permissions and the ability to isolate local Model Context Protocol (MCP) servers in Podman containers. This setup provides control over personal and sensitive workflows, allowing businesses to connect their AI personas with private data and various external APIs while keeping operations transparent. By allowing tailored usage of LLMs based on their strengths, such as using ChatGPT for analysis or Claude for writing, StrictLLM is set to optimize AI performance and usability for developers, organizations, and power users alike.