Knowns CLI

The third feature group includes templates and AI agent context via MCP. Templates allow users to capture proven workflows—such as research, planning, implementation, and pattern extraction—and reuse them across projects. These are accessed through the `/kn-*` namespace in tools like Claude Code, making complex tasks repeatable and consistent. The Agent Context feature uses the Model Context Protocol to provide AI assistants with structured project memory automatically. Instead of pasting context manually, the AI reads the full state of tasks, docs, and decisions, enabling it to act with complete awareness. This integration means that every AI session is grounded in real project data, reducing hallucinations and improving output quality. Templates also serve as a form of knowledge sharing, allowing teams to standardize their approaches. The MCP integration is a key differentiator, as it seamlessly connects Knowns CLI to the AI ecosystem.

The workflow of Knowns CLI follows a simple five-step cycle: Capture, Link, Work, Verify, and Remember. In the Capture phase, users write down what needs to happen and clearly define the definition of done. Next, they Link these tasks to relevant docs, specs, templates, and past decisions. During the Work phase, the AI reads the full context and follows the plan. After implementation, the Verify step checks acceptance criteria before marking the task complete. Finally, the Remember phase extracts patterns and decisions for future reuse, closing the loop. This cycle ensures that knowledge is continuously captured and applied, preventing context from being lost between sessions. The structured approach makes it easy to see the status of each task and the rationale behind every decision, fostering transparency and accountability.

Concrete use cases include a solo developer working on a Flutter dark mode implementation across multiple sessions. By using Knowns CLI skills like `/kn-research`, `/kn-plan`, `/kn-implement`, and `/kn-extract`, the AI maintains context of the project state—what's done, what's next, and past decisions. Another use case is an AI-assisted team where multiple agents collaborate on the same codebase. Knowns ensures everyone works from the same structured context, preventing conflicting assumptions. In product-engineering handoffs, the full specification, linked docs, and acceptance criteria are automatically available, reducing onboarding time and ensuring alignment. These scenarios demonstrate how Knowns CLI reduces friction in modern development workflows. The outcome is faster delivery, fewer errors, and less cognitive load for team members. By keeping context intact, Knowns CLI enables a seamless handoff between sessions and between people and AI.

Target users include solo developers who build projects across multiple sessions, AI-assisted teams with multiple agents and humans, and product-engineering handoff scenarios. The tool is platform-agnostic and can be installed via Homebrew, shell script, PowerShell, npm, or npx, requiring Node.js 20+. It is open source under the MIT license with no vendor lock-in, available at GitHub. The tool also offers a Web UI with kanban, docs, and chat capabilities, but its core strength lies in the CLI and MCP integration. Pricing is free as open source, with no hidden costs. Whether you are a solo developer or part of a large team, Knowns CLI helps you stop re-explaining and start building with full context. In summary, Knowns CLI provides persistent AI project memory that connects tasks, docs, and decisions in one searchable system, making it indispensable for anyone seeking to maintain project context across sessions and AI interactions.