AI search engine and MCP server for bots and agents. One Rust binary indexes code, docs, logs, JSONL/CSV and PDFs for BM25 full-text, vector and hybrid search, code search without grep, agent memory and RAG. Elasticsearch-compatible API; Corpus Hub serves hundreds of reference memory corpora.

⚡ DocsAgent — give your AI agents instant, private access to your personal knowledge base (Zotero, Obsidian, Apple Notes supported now, local docs on the road). Native C++ search core: BM25 + passage ranking in ~15 ms. MCP server for Claude, Cursor, Cline & any MCP client.

Semantic code search plugin for Claude Code using hybrid vector search + BM25. 98% accuracy, 5x faster than grep.

Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.

⚡ Cut Claude token usage by 90%+ — free, open-source, local-first context compression for Claude Code. Hybrid RAG (BM25 + ONNX vectors), AST chunking, reranking. No API needed.

MCP plugin that saves 98% of your context window. Works with Claude Code, Gemini CLI, VS Code Copilot, OpenCode, and Codex CLI. Sandboxed code execution, FTS5 knowledge base, and intent-driven search.

Cross-session memory plugin for DeepSeek Harness: seven-layer SQLite store (soul/user/project/fact/lesson/topic/rules), BM25 retrieval, per-window dream consolidation. 跨会话七层长期记忆插件。

Self-learning vector memory for AI agents — single-file .rvf cognitive container with HNSW search, episodic Reflexion memory, causal graph + Cypher, 9 RL algorithms, Thompson Sampling bandit, 41 MCP tools, hybrid (BM25 + dense) retrieval, GNN attention. 1

MCP server giving AI agents (Claude Code, Claude Desktop, Cursor, ChatGPT, Codex, OpenClaw) persistent long-term memory backed by your local Obsidian markdown vault. Hybrid retrieval (BM25 + ML embeddings + BGE reranker, RRF-fused), HNSW + int8 quantizati

Enables AI assistants to persist and recall local memory across sessions and projects through hybrid BM25/TF-IDF retrieval over a file-based vault. Facts are tracked with dual-clock timestamps and require no native dependencies.

Self-hosted semantic code search platform — Go server with web dashboard, CLI, and AI-agent skills. Search code by meaning, not text: hybrid BM25 + dense embeddings via llama.cpp.

Context engineering for AI agents. ~80% fewer tokens. Fix tool overload. Skills and memory with in-process BM25 and semantic retrieval. Progressive Disclosure. No vector DB.

🧠 The memory that dreams — cross-session memory for DeepSeek Harness. Local SQLite + Markdown mirrors, autoDream consolidation, 4-way recall (keyword/vector/BM25/entity), and a memory panel. 10 model tools, MCP server, CLI.

Self-hosted memory service for AI agents that work in teams, written in Rust: layered L0→L3 memory, async LLM distillation, hybrid BM25 + vector recall, and department/team/member isolation. SQLite or Postgres, any OpenAI-compatible LLM.

MCP server giving AI agents (Claude Code, Claude Desktop, Cursor, ChatGPT, Codex, OpenClaw) persistent long-term memory backed by your local Obsidian markdown vault. Hybrid retrieval (BM25 + ML embeddings + BGE reranker, RRF-fused), HNSW + int8 quantizati

Enables Claude to search and read an organization's internal documents (Word, PDF, HTML, Markdown) through hybrid BM25 + vector retrieval, returning answers with exact citations while enforcing server-fixed role permissions, hiding confidential and superseded content, and logging the token cost of every request.

Enables users to index and retrieve local documents through semantic, BM25, or hybrid search, with tools for chunk reads and page images accessible from MCP clients such as Codex and ChatGPT web.

Enables AI agents to store decisions as a versioned history, linking each new decision to the one it replaces so that recall always resolves to the decision currently in force instead of obsolete ones. It runs fully locally on a single SQLite file with BM25 search, and can optionally detect replacements automatically via an LLM or heuristic linker.