Search

TurboQuant compression for LLM KV caches, pgvector embeddings, and NATS transport — 5-10x memory reduction

Terminal client for OpenAI chat completions

Typed envelope and telemetry primitives for LLM cost attribution with OpenTelemetry instrumentation

Allows renegotiating payment installments on posted invoices in Brazil

A high-level NLP toolkit built on top of modern LLMs.

LLM Cost Intelligence SDK — track, optimize, and control your AI spend

SUNWÆE gen — multi-provider LLM engine library.

KV cache fingerprinting for persistent cross-session LLM memory. Fourier decomposition achieves 98% Recall@1 at 51µs.

Compressed vector and graph-augmented retrieval engine with adaptive two-stage search, implementing TurboQuant/QJL quantization from Google Research.

Rust-powered LLM essay grading engine

A comprehensive RAG framework with Arabic language support, multi-provider embeddings, and advanced RAG strategies

Codebase health scanner and Claude Code skill. Finds dead code, unused imports, duplicated functions, hardcoded secrets, circular dependencies. 20 languages.

AI-powered video podcast creation skill for coding agents. Supports Bilibili & YouTube, multi-language (zh-CN/en-US), 6 TTS engines (Edge/Azure/ElevenLabs/OpenAI/Doubao/CosyVoice), 4K Remotion rendering.

LLM chat lab project

Claude Code skill that will generate interactive documentation for your project

Ticket Takeaway - a control panel to help manage multiple agents directly via cli

Skills for working on ASU's SOL supercomputer

Marketing skills for Claude Code — SEO audits and implementation, ad analysis, ad optimization. Free skills need no account. $9/mo for live Search Console, Bing & LinkedIn data.

Clean-room AI coding agent in Python, inspired by modern systems like Claude Code.

The Next-Gen Agent-Native Skill Recommendation Engine