Vector infrastructure and Matryoshka embeddings for AI agents.
Free Claude Code access via Matryoshka Studio. No Anthropic account needed.
Recursive Language Model - Process documents larger than LLM context windows
Generate your file like a Matryoshka
Zero-dependency vector store for Node.js and browser. Float32, Int8, and 1-bit quantization with IVF indexing and Matryoshka search.
Create a mock API Gateway by rewrite your existing APIs with mock fields and generating new API on the fly.
Recursive Prettier plugin for embedded code formatting across nested multi-language blocks.
Vercel AI SDK provider for Forge — Voxell's text-embedding API. turbo/pro/ultra (ultra = Qwen3-Embedding-8B; ~75+ avg MTEB, #4 English, top usable model), Matryoshka dims.
Recursive Language Model - Process documents larger than LLM context windows
MCP server for Forge — Voxell's text-embedding API (turbo→ultra; ultra = Qwen3-Embedding-8B, ~75+ avg MTEB). embed + list_models for semantic search & RAG.
CLI tool to visualize Node.js dependency chains as trees
Reusable VM sandbox with REPL state persistence, session management, FSM engine, and correction loop
Core RAG engine for RagClaw - extractors, chunkers, embedder, store
An extensible MCP server that enhances coding productivity with AI-powered features including semantic code search, intelligent indexing, and more, using local LLMs
just-bash plugin: db (MongoDB-style document store) and vec (vector similarity search) commands for in-shell agents.
Basic utilities for Vue and Vuex
SIMD-accelerated MaxSim (ColBERT/ColPali), cosine similarity, diversity (MMR/DPP), token alignment/highlighting for vector search and RAG. Supports text and multimodal late interaction.
CLI for RagClaw - local-first RAG engine
Pocket-Sized Multimodal AI for Content Understanding and Generation
High-performance, zero-RAM structural data collection engine for Node.js.
ptv vissim crack [patched] 57
Official SDK for memory.automators.work — sovereign memory as an API for AI agents. Types generated from the live OpenAPI 3.1 spec.
Provides a unified API for both file system access and access to ZIP archives.
Lexa CLI: hybrid local search (BM25 + binary-quantized Matryoshka KNN + cross-encoder rerank) over arbitrary file trees. `lexa index <path>`, `lexa search <query>`, `lexa watch <path>`.
Local-first hybrid retrieval engine: BM25 + binary-quantized Matryoshka KNN + cross-encoder rerank, in one Rust crate. Pairs with `lexa-obsidian` for vault-aware MCP.
SIMD-accelerated MaxSim, cosine, diversity (MMR/DPP) for vector search and RAG pipelines
SIMD-accelerated vector similarity primitives with binary, ternary, and scalar quantization
Rank fusion and reranking: RRF, ISR, CombMNZ, Borda, DBSF, MaxSim (ColBERT), MMR, DPP, Matryoshka.
rmcp stdio MCP server for the Lexa hybrid retrieval engine. Exposes `search_files`, `index_path`, `list_indexed_paths`, `purge_path`, and `status` to any MCP client (Codex, Claude Desktop, Cursor, Claude Code) for arbitrary local file trees. Pairs with `lexa-obsidian-mcp` for vault-shaped queries.
AI agent memory with local SQLite backend — FTS5, fuzzy, and vector search
Embedding interfaces + local backends (Candle/HF).
IMECE v0.1.0 - Local-First Autonomous Agent Framework. Zero API dependency. Zero cloud lock-in. Full sovereignty.
A package to query popular LLMs
Multi-paradigm embedding library: ColBERT, dense, sparse, vision-language, and time series models
Matryoshka is a system for parsing html templates. It merges files based on their position in directories.
Helps you spawn new inner views and lookup the right one.
Russian dolling cache style in redis
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.