NeuroAIgent sums up the idea behind it in one line: “The AI doesn't own the memory. You do.”
How does Central Brain AI work?
Central Brain by NeuroAIgent is a local-first AI memory layer for the AI tools you already use. The desktop app for Windows and Apple silicon Macs prepares and indexes project files on storage you choose, using a bundled 384-dimension MiniLM embedding model and a local index, and serves project memory and source-traceable search to supported AI tools over the Model Context Protocol (MCP). Memory is kept as readable local files: five note types (decision, fact, question, todo and summary) plus a project brief. Licensing uses Keygen; connected AI tools receive the context they retrieve, and optional cloud connections send selected data only when you enable them.
Memory. Five note types (decision, fact, question, todo and summary) plus a project brief, stored as readable local files.
MCP tools. Eight MCP tools. File tools are read-only; memory_checkpoint appends notes; memory_compact replaces the project brief and marks older notes as compacted rather than deleting them. On connect, central_brain_ping returns a RESUME block from the project’s saved memory, so a new session picks up where the last one left off.
central_brain_pingmemory_recallmemory_checkpointmemory_compactsemantic_searchsearch_textread_filelist_files
Which AI tools work with Central Brain AI?
Claude Desktop, Claude Code, GitHub Copilot and the ChatGPT desktop app (Chat, Work and Codex modes) on Windows and Apple silicon Macs. Capabilities vary by client.
| AI tool | Platforms |
|---|---|
| Claude Desktop | Windows, macOS (Apple silicon) |
| Claude Code | Windows, macOS (Apple silicon) |
| GitHub Copilot | Windows, macOS (Apple silicon) |
| ChatGPT desktop app (Chat, Work and Codex modes) | Windows, macOS (Apple silicon) |
For setup steps, see the MCP memory server guide.
What stays on your computer?
In the default workflow, files, prepared output, vectors and memory stay on storage you choose. Keygen receives licensing and device data, not documents. Connected AI tools receive the context they retrieve, and optional cloud connections send selected data only when enabled.
Central Brain runs air-gapped: preparing files, indexing, search and memory need no network connection. The license is re-checked online every 14 days, and Team and Enterprise plans can get a build with no online license check. Optional cloud connections and cloud-hosted AI clients need internet.
What does Central Brain AI index?
Central Brain prepares and indexes PDF, DOCX, TXT, Markdown, CSV and XLSX, with metadata-only indexing for images, audio and video; OCR is optional and depends on the build.
Source traceability. Results include the source path, prepared-file path, page and chunk where available.
Embeddings. A bundled 384-dimension MiniLM model and a local JSONL index; cloud embeddings are optional.
Is Central Brain AI related to the “central brain” in neuroscience?
No. Central Brain is NeuroAIgent's desktop AI memory software, not a chatbot. It is unrelated to the anatomical central brain and the Central Brain Tumor Registry of the United States, and is not affiliated with other projects, products or services that share the name.
How does Central Brain AI compare with other AI memory tools?
Central Brain is a local-first AI memory layer: it keeps memory on storage you choose and serves it to several supported AI tools over MCP, instead of keeping memory inside one assistant or inside a service that developers connect to their own applications.
| Kind of AI memory | Where memory lives | Typical fit |
|---|---|---|
| Memory built into one assistant | Inside that provider's product and account | Personalization within that assistant |
| Developer memory service or framework | In a service or database that an application reaches through an API or SDK | Teams building memory into their own application |
| Central Brain | Readable local files and a local index on storage you choose | Reusing project context across sessions and supported AI tools |
These categories overlap, and products change. Compare the data boundary, the supported clients, how you inspect and correct memory, and what needs a network connection, rather than the word “memory” alone. The AI memory guide explains the category in more depth.
Common questions
Is “Central Brain AI” the same as Central Brain by NeuroAIgent?
Yes. Central Brain AI is another name for Central Brain, NeuroAIgent's desktop AI memory software. It is also called Central Brain by NeuroAIgent and NeuroAIgent Central Brain.
Is Central Brain by NeuroAIgent related to other projects called Central Brain?
No. Central Brain is NeuroAIgent's desktop AI memory software, not a chatbot. It is not affiliated with other projects, products or services that share the name.
Who is Central Brain AI for?
Central Brain is built for developers, consultants, researchers and small teams who reuse project files across AI sessions.
Which operating systems does Central Brain AI run on?
Central Brain is a desktop app for Windows and macOS (Apple silicon). The getting-started guide has current setup and download information.
How is Central Brain AI licensed?
Central Brain uses Keygen licensing, one license per computer, re-checked online every 14 days; Team and Enterprise plans can get a build with no online license check. See Central Brain pricing for current plans.
Does Central Brain AI need an internet connection?
Central Brain runs air-gapped: preparing files, indexing, search and memory need no network connection. The license is re-checked online every 14 days, and Team and Enterprise plans can get a build with no online license check. Optional cloud connections and cloud-hosted AI clients need internet.
Next steps
- Central Brain product page: the canonical product description
- Central Brain pricing: current plans
- MCP memory server guide: connect a supported AI tool over MCP
- What is an AI memory layer?: the category, explained