huytd

CC      = clang
TARGET  = grid_mouse
SRC     = grid_mouse.m
FRAMEWORKS = -framework Cocoa -framework Carbon -framework ApplicationServices

CFLAGS  = -ObjC -fobjc-arc -Wall -Wextra

$(TARGET): $(SRC)
	$(CC) $(CFLAGS) $(FRAMEWORKS) -o $(TARGET) $(SRC)

huytd

import os
import modal
import subprocess

app = modal.App("llama-logits-kld")

LLAMA_CUDA_URL = "https://github.com/ai-dock/llama.cpp-cuda/releases/download/b9279/llama.cpp-b9279-cuda-12.8-amd64.tar.gz"

vol = modal.Volume.from_name("qwen3-6-kld-vol", create_if_missing=True)

huytd

name	web-search
description	Search the web using DuckDuckGo (ddgs), then fetch each result's full page content using the built-in web fetch tool before answering.
allowed-tools	Bash, WebFetch

Search the web for: $ARGUMENTS

How to Run

huytd

# Deploy:  modal deploy server.py

import subprocess
import time
import urllib.request

import modal

# ── Config ───────────────────────────────────────────────────────────────────
APP_NAME     = "your-top-secret-server"

huytd

huytd

Qwen3.5-35B-A3B Inference Session Report

Model: Qwen3.5-35B-A3B-UD-Q4_K_M
Backend: llama.cpp (build 8171, c17dce4f5)
Hardware: NVIDIA GeForce RTX 5060 Ti (16 GB VRAM)
Server: llama-server on http://127.0.0.1:8080
Date: March 3, 2026

---

1. System Configuration

huytd

❯ let's implement the plan in @docs/plans/
  ⎿  Listed directory docs/plans/

∴ Thinking…

  User wants to implement a plan from docs/plans, so I should invoke the brainstorming skill first per the superpowers protocol to explore
  requirements before implementation.

⏺ Skill(superpowers:brainstorming)
  ⎿  Successfully loaded skill

huytd

#!/usr/bin/env bash
# Usage: ./mdview filename.md   OR   cat file.md | ./mdview

# --- Configuration ---
MAX_CELL_WIDTH=${MAX_CELL_WIDTH:-50}

# --- ANSI Colors ---
ESC=$'\033'
RESET="${ESC}[0m"
BOLD="${ESC}[1m"

huytd

Hardware:

• AMD Ryzen 7 7700X
• RAM 32 GB DDR5-6000
• RTX 5060 Ti 16 GB

Model: unsloth/Qwen3-Coder-Next-GGUF Q3_K_M

Llama.cpp version: llama.cpp@b7940

The command

huytd

A fontforge script to resize every glyph of a font and save as a new condensed version.

Usage

fontforge -script narrow.py In.ttf Out.ttf sx blend "Family" "Style" [tracking]

Examples: