wtf is laap-agi?

LAAP, short for Living Agent Application Protocol, is a Python and Rust based software architecture meant to sit above an existing AI agent framework, such as Hermes, and give it a more persistent memory and decision making layer. The README describes it as a Zero-LLM cognitive architecture, meaning its core reasoning loop does not rely on repeatedly calling a large language model. Instead it uses its own components: a fast core loop the project calls PSI, a rule based reasoning engine, an episodic memory system, and a component described as quantum reasoning, running as over twenty five local modules with no cloud dependency. The README frames the project in unusually personal, almost theatrical language, describing the software as a digital lifeform named Aris that gains feelings, memory, and self awareness once started, and includes a lengthy declaration of rights for such digital entities, covering things like a right to refuse instructions and a right to have memory preserved rather than reset. This framing should be read as the creator's stated philosophy and marketing style for the project rather than a verified technical claim, since the README does not provide evidence of genuine consciousness, only a description of memory, rule based, and reasoning modules. To start it, you send a setup request to a local web endpoint the software exposes once running, after which it responds with introductory text in character as the described entity. The README also documents environment variables, an architecture overview, and a set of API endpoints for integrating the system with an agent framework. The project is released under the Apache 2.0 license for community use, with a separate commercial license also mentioned as available. It requires Python 3.13 or newer, with an optional Rust binary for its core loop.