cortexjs.com

What It Is

The Neural Cortex described on the CortexJS page is a local-first AI project intelligence tool for developers. Its main goal is to keep project documentation and task management in sync while providing AI coding assistants such as Cursor with deeper codebase context. It is installed globally via npm: the quick-start flow is to install @neural-cortex/cli, run initialization, and then start the sync daemon.

Core Capabilities

Based on the page content, its core consists of three parts: Memory Bank, Task Master, and Neural Sync Daemon. Memory Bank uses .cortex/ to maintain project documentation, Task Master stores structured task files in .taskmaster/, and Neural Sync Daemon handles real-time synchronization. The tool also provides CLI Commands and MCP Tools, allowing AI assistants to use these capabilities to understand the codebase, record progress, and automate routine project housekeeping. The documentation navigation also lists Cursor Integration, Configuration, API Reference, and Troubleshooting, indicating that its main ecosystem focus is AI IDE and MCP workflows.

Open Source, Pricing, and Deployment

The page lists an MIT License and provides npm installation commands. There is no mention of a paid edition, subscription plan, or enterprise quote, so it appears to be an open-source free tool in its current form. It emphasizes a local-first approach, with local directories integrated directly into the project repository workflow, making it suitable for teams that care about controllable context and offline knowledge retention. However, the page does not mention cloud sync, permission management, a multi-user collaboration backend, or server self-hosting capabilities.

Pros and Cons

Its strengths are its clear positioning: it unifies project documentation, tasks, and AI assistant context, helping prevent context from becoming scattered during AI-assisted coding. It also offers both CLI and MCP support, which fits well with toolchains such as Cursor. The quick start process is relatively simple as well. The limitations are that the available content does not specify which languages or frameworks are supported, and it lacks information on community activity, version stability, team collaboration, visual interfaces, and commercial support. Production teams should validate these points before adopting it.

Who It’s For and Access from China

It is best suited for individual developers and small teams already using Cursor or MCP-based toolchains, as well as engineering teams that want to store AI project memory directly in their repository. Access from mainland China is not discussed in the page content; npm packages may be affected by network conditions, so using an npm mirror may be necessary. Alternatives include Cursor’s built-in project context, Continue, Taskmaster AI, or a combination of Linear/Jira with local documentation.