@XAMTO_AI: Let AI stay up late to modify code while you sleep. This tool is called gnhf, short for "Good Night, Have Fun" — in plain English: you go rest, leave the rest to AI. The workflow is quite simple: ① Write a single command, and it automatically invokes Claude Code or Cod…
Summary
gnhf is an open-source tool that lets AI agents automatically loop through code modifications, generate Git commits, handle errors, and record progress while you sleep. It supports multiple AI models (e.g., Claude Code, Codex) and multi-agent parallel work.
View Cached Full Text
Cached at: 05/22/26, 07:58 PM
Before I go to bed, I tell my agents:
good night, have fun
Similar Articles
@Xudong07452910: ARIS: The magic tool that lets AI keep doing research while you sleep! An ultra-lightweight automated research tool that allows Claude Code / Codex / Cursor / Trae / domestic models to automatically enter the research workflow: read papers, find weaknesses, generate ideas, …
ARIS is a lightweight automated research tool based on Markdown that supports multiple AI models (such as Claude Code, Cursor, Trae) to automatically complete the entire research workflow including reading papers, generating ideas, designing experiments, writing papers, etc., allowing AI to continuously explore while users sleep.
@XAMTO_AI: Previously, with AI coding, I was always in a 'human-supervising-tool' mode—completing one task before moving to the next, like an assembly line worker. Multica has revolutionized this collaboration approach. It's not merely a code generator that requires constant prompting, but a virtual colleague engineered to autonomously accept tasks and proactively report progress. You simply need to on the kanban...
Multica is an open-source workspace designed to transform AI coding agents into virtual colleagues that can autonomously accept tasks and report proactively, thereby changing how developers collaborate with AI tools.
@Ryrenz: Most people using coding AI just say “help me fix this bug,” then spend half an hour wrestling with it. Change the phrasing and it gets it right the first time. 1. Have it look before acting “Don't write code yet. Read the relevant files and tell me how this feature currently works; I'll confirm before you change it.” 2. Stuck bug “Don't guess. Add log…
A practical tip on how to use coding AI more effectively. It lists 12 concrete prompt examples—such as having the AI read the code before modifying it, changing only the necessary parts, and writing rules into AGENTS.md—to reduce back-and-forth and make the rules stick long-term.
@ai_super_niko: AI coding tools execute frantically, but you can only watch helplessly as they run in the wrong direction? Made a mistake in the previous query and have to add ten explanations afterward? I summarized my recent heavy usage of Codex into four actions: 1. Interrupt: double-tap Esc to stop on the spot, refusing to wait idly; 2. Edit: directly edit the latest message...
Summarizes four key actions for using the AI coding tool Codex: interrupt, edit, guide and queue, fork — helping users transition from passively waiting for results to actively managing the process.
@Saccc_c: A few tips for beginners to quickly get started with AI: 1. Stop using conversational AI like ChatGPT and start using Agent tools that can autonomously execute tasks, such as Claude Code, Codex, Manus. 2. Write your own Skill, for example, structuring your work habits, preferences, …
Tips for AI beginners: Use Agent tools that can autonomously execute tasks (e.g., Claude Code, Codex, Manus), write a structured personal Skill description, and continuously accumulate context to refine effective prompts.