[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"readme:spector":3},"\u003Cp align=\"center\">\n  \u003Cpicture>\n    \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\"docs\u002Fassets\u002Fspector-logo-full-dark.png\">\u003C\u002Fsource>\n    \u003Csource media=\"(prefers-color-scheme: light)\" srcset=\"docs\u002Fassets\u002Fspector-logo-full-light.png\">\u003C\u002Fsource>\n    \u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fspectrayan\u002Fspector\u002FHEAD\u002Fdocs\u002Fassets\u002Fspector-logo-full-dark.png\" alt=\"Spector\" width=\"600\" \u002F>\n  \u003C\u002Fpicture>\n\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Cstrong>The Zero-Overhead, Agent-Ready AI Memory Engine.\u003C\u002Fstrong>\n\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fspectrayan\u002Fspector\u002Fblob\u002FHEAD\u002FLICENSE\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache_2.0-blue.svg?style=for-the-badge\" alt=\"License\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fopenjdk.org\u002F\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FJava-25-orange.svg?style=for-the-badge\" alt=\"Java\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fpypi.org\u002Fproject\u002Fspector-client\u002F\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fspector-client?color=3776AB&amp;style=for-the-badge&amp;logo=pypi&amp;logoColor=white\" alt=\"PyPI\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002F@spectrayan\u002Fspector-client\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fv\u002F@spectrayan\u002Fspector-client?color=CB3837&amp;style=for-the-badge&amp;logo=npm&amp;logoColor=white\" alt=\"npm\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fspectrayan\u002Fspector\u002Fpkgs\u002Fcontainer\u002Fspector\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDocker-GHCR-2496ED?style=for-the-badge&amp;logo=docker&amp;logoColor=white\" alt=\"Docker\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fspectrayan\u002Fspector\u002Factions\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Factions\u002Fworkflow\u002Fstatus\u002Fspectrayan\u002Fspector\u002Fci.yml?branch=main&amp;style=for-the-badge\" alt=\"Build\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fspectrayan.github.io\u002Fspector\u002F\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDocs-MkDocs-blue?logo=materialformkdocs&amp;style=for-the-badge\" alt=\"Docs\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fspectrayan\u002Fspector\u002Fblob\u002FHEAD\u002FACKNOWLEDGMENTS.md#open-source-contributors\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcontributors\u002Fspectrayan\u002Fspector?style=for-the-badge&amp;color=blue\" alt=\"Contributors\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fwww.bestpractices.dev\u002Fprojects\u002F14829\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fwww.bestpractices.dev\u002Fprojects\u002F14829\u002Fbadge\" alt=\"OpenSSF Best Practices\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Cem>\"A database returns what was written. A memory engine reconstructs what is reachable from a cue — under decay, association, and tier physics.\"\u003C\u002Fem>\n  \u003Cbr \u002F>\n  \u003Csub>— \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fspectrayan\u002Fmemory-fundamentals\" rel=\"nofollow ugc noopener\">Memory Fundamentals MF-001\u003C\u002Fa>\u003C\u002Fsub>\n\u003C\u002Fp>\u003Chr \u002F>\n\u003Cp>Legacy AI stacks bolt memory onto stateless vector databases — storage without cognition. \u003Cstrong>Spector\u003C\u002Fstrong> is a cognitive memory engine for AI agents: it remembers, forgets, consolidates, and forms associations across working, episodic, semantic, and procedural tiers, then retrieves with fused semantic and hybrid scoring at sub-millisecond latency. Traces are linked by co-activation, temporal, and entity edges, so recall can surface what is related, not only what matches a vector. Connect any agent through the built-in \u003Cstrong>MCP server\u003C\u002Fstrong>, call it over \u003Cstrong>REST\u002FgRPC\u003C\u002Fstrong>, drive it from the \u003Cstrong>Python or TypeScript SDKs\u003C\u002Fstrong>, or embed it directly in the JVM. Every user, agent, or tenant is physically isolated in its own on-disk namespace — true data separation, not a shared-store filter. Under the hood, Java Project Panama and the Vector API deliver SIMD scoring with measured near-zero garbage-collection pressure, and no external database.\u003C\u002Fp>\n\u003Chr \u002F>\n\u003Ch2>⚡ 30-Second Quickstart\u003C\u002Fh2>\n\u003Cp>Connect an agent, install an SDK, or launch a local node in seconds:\u003C\u002Fp>\n\u003Ch3>1. Zero-Install MCP Server (for AI Agents)\u003C\u002Fh3>\n\u003Cp>Run instantly via NPX — connects to a running local Synapse daemon on \u003Ccode>:7070\u003C\u002Fcode> if healthy, or runs an embedded memory kernel (requires OpenJDK 25+):\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-bash\"># Agent runner: connects to local daemon or launches in-process MCP kernel\nnpx -y @spectrayan\u002Fspector mcp\n\n# Or build and launch from source:\nmvn clean package -pl synapse\u002Fspector-cli -am -DskipTests\njava --enable-preview --add-modules=jdk.incubator.vector -jar synapse\u002Fspector-cli\u002Ftarget\u002Fspector.jar mcp\n\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Ch3>2. Client SDKs (Zero Java Required)\u003C\u002Fh3>\n\u003Cp>Interact with Spector over HTTP \u002F SSE from your language of choice:\u003C\u002Fp>\n\u003Cp>\u003Cstrong>Python:\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-bash\">pip instal\n\u003C\u002Fcode>\u003C\u002Fpre>\n",1795123406891]