Search

AI-powered Linux log analyzer with a terminal UI

A fast, polished CLI for picking and launching Ollama agents and models interactively

ArquiSys AI Studio - One command to run everything (Sprint 3: Agente Validador + LLM Generators)

A pi-based coding agent optimized for small local language models. Reproduces the whitepaper's scaffold-model-fit adaptations as pi extensions.

Deep Code CLI - Vibe coding for the deepseek-v4 model in your terminal

Tau - multi-provider AI coding CLI with OpenAI, Gemini, Groq, DeepSeek, Ollama and more.

Manage Claude Code API subscriptions - switch between different API providers and models

Supabase adapter package for `blog-kit`.

Vlix web app and desktop agent connector.

Lightweight SQL registry with named parameters

Redivo AI provider plugin for OpenCode — use GPT-5.3, GPT-5.4, GPT-5.5, and more via your Redivo AI account

Deterministic prose style metrics for measuring variety, density, repetition, and narrative texture. Designed for AI editorial pipelines and writing tools.

REFRACT — Reference-anchored Robust Acid-test for Compressed Transformers. Multi-axis KV-cache fidelity scoring for LLMs across llama.cpp, MLX, vLLM, and SGLang.

The Node framework that owns its stack.


Cartão de apresentação do Carlos Mesquita no terminal — rode com `npx carlos3g`.

Embed signed semantic-metadata layers into images, PDFs, and audio files

Zero-dependency visual overlay for data-ref element labels. Three label modes, depth control, outline guides, click-to-copy. Default hidden — press H to reveal. Works in static HTML, Vite/Webpack builds, React/Next/Vue, WordPress, and Shopify.

collection of test helping functions

She does all the work so you don't have to. Watches GitHub issues and spawns Claude Code sessions to work them.