⚡ Spector — The AI Memory Backbone¶

Zero-overhead, agent-ready AI search and cognitive memory — embedded in a single JVM.

Spector is a Java-native AI search engine and cognitive memory system that combines SIMD-accelerated vector search, keyword search (BM25), and biologically-inspired memory consolidation into a single embeddable library. No Docker, no external databases, no Python — just a JAR.

Connect AI agents via the built-in MCP server (Claude Desktop, Cursor, custom agents), embed directly in your Spring Boot app, or run standalone. Spector delivers sub-millisecond search at scale with zero garbage collection pressure thanks to Project Panama off-heap memory.

🔥 Key Numbers¶

🗺️ Choose Your Path¶

💡 How It Works¶

Spector combines three search modalities — semantic vectors, keyword matching, and cognitive scoring — into a single fused pipeline:

graph LR
    A["🤖 AI Agent"] --> B["📡 MCP Server"]
    B --> C["⚡ SpectorEngine"]
    C --> D["🧠 Hybrid Search"]
    D --> E["🎯 RRF Fusion"]
    E --> F["🤖 LLM Re-ranking"]
    F --> G["✨ Results"]

    H["📄 Document"] --> I["🧩 Chunking"]
    I --> J["🧬 Embedding"]
    J --> C

What Makes Spector Different¶

• Embedded deployment — runs as a library inside your JVM. No Docker, no servers, no network hops.
• Agent-native — 13 MCP tools for search, memory, and cognitive operations. Connect Claude Desktop or Cursor in one config line.
• Cognitive memory — the only system combining power-law decay, Two-Factor strengthening (Bjork & Bjork), emotional valence, and Hebbian association in a single scoring formula.
• Zero GC pressure — all vector data and headers live off-heap via Project Panama. The JVM garbage collector never sees memory records.
• SIMD everywhere — vector distance, quantization, and scoring use Java Vector API (AVX2/AVX-512/NEON) for hardware-accelerated computation.

🌟 Project Stats¶

Built with ⚡ by Spectrayan · GitHub · Apache 2.0 · BSL 1.1 (memory)