Show HN: The TERSE state language and protocol A developer has released TERSE, a pre-release (alpha) unified state language and protocol designed for efficient state storage, querying, and mutation, with a focus on simplicity and token efficiency. The project includes a reference Python implementation, an MCP server, a memory module, and a Karpathy-brain compatible API that claims 1/6th token use and 1/8th the number of tool calls compared to alternatives. Hello World valid terse; for Show HN I'm terse greetings Container "A container may have one text block, single or multi-line" Object Semi-colon separated attributes; variables like: 123; "Traditional unicode strings allowed here too." Another object with a name multiple objects under a container Further description of TERSE """ TERSE is a unified way to store state, with a defined way to query and mutate it that has been relentlessly refined for overall simplicity and token efficiency not just at rest, but in operation. We realized state access was a low-level, common pattern that keeps getting reinvented. It goes deeper than just AI memory. Consider that Anthropic memory and others are a bunch of markdown files and in some respects, they got it right... simpler is better. So REALLY the problem is settling on a format that is flexible to cover 95% of a domain. We can do so MUCH better than Markdown; TERSE is opinionated on that but it has very good reasons. By being a line-ordered and line-identifiable structure with special treatment for free-form blocks like this , now the whole state can be idempotently re-declared. Literally, this text of this post is a TERSE state declaration that an AI or you could make. Or, duh, just copy this to a .terse file So let's say you did that. To query this state: ? Hello World To copy this whole state somewhere: My new container ? Hello World //inline queries physically expand into the declaration Directives are a tail with composable set of operations. For example, let's clean that up.. I declare thee removed My new container REMOVED // all contents too There's more. TERSE makes it exceeding easy for the AI to do stuff with state. The MCP has one main op where these two things are sent together 1 state declarations 2 state queries post-application of declarations and results unioned BTW, to query all state, wait for it... ? // Not that you want do this often TERSE has a full object model for de serialization. A reference Python implementation of the full specification is also included. """ Example apps included to help get you started; in the monorepo TERSE MCP "store anything you want locally for your AI" TERSE Memory a generalist; modify your own TERSE Brain "A Karpathy-brain compatible API but with TERSE as the backing store. 1/6th token use and 1/8th the number of tool calls " Misc TERSE Browser query-response UI; great for testing VS Code syntax highlighter vsix included Reminders Terse is pre-release alpha Comments URL: https://news.ycombinator.com/item?id=49045760 https://news.ycombinator.com/item?id=49045760 Points: 1 Comments: 1