# SamuelQZQ/auto-coding-agent-demo

一个长达 10 小时的 AI 全自动编程实验

Repository: https://github.com/SamuelQZQ/auto-coding-agent-demo
Canonical: https://ross.abutalabs.com/products/auto-coding-agent-demo
Language: TypeScript
License Family: other
Last push: 2026-02-13T19:25:47+00:00

## Health v2 (maintenance only)
Score: 45/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 67, release rhythm 35, longevity 14
- inputs: {"age_days": 201, "days_push": 201, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1027, forks 386 (observed 2026-08-28T04:03:17.178107+00:00)

## What it is
A demo repository documenting a 10-hour experiment where an AI coding agent (Claude Code) autonomously wrote 100% of the code for a Next.js AI video generation app. It includes the workflow definition (CLAUDE.md), task list (task.), and scripts for running long autonomous coding sessions.

## Use cases
- run an AI agent that codes autonomously for hours
- learn how to structure tasks for long-running AI coding agents
- generate a full-stack Next.js app with Claude Code
- set up an automated loop for AI-driven development
- study how AI-generated code evolves through git history
- build an AI video generation app as a coding agent testbed

## When to choose
- you want to experiment with or learn long-running autonomous AI coding workflows with Claude Code
- you need a template for task-driven AI development with CLAUDE.md and task.
- you want a reference example of a fully AI-generated full-stack project

## When to avoid
- you need production-ready, human-reviewed software with a license and maintenance guarantees
- you don't use Claude Code or want a vendor-neutral agent framework
- you require guaranteed safety, since the demo encourages permission-skipping autonomous execution

## Facets
- artifact type: learning-resource
- maturity: experimental
- function: agent-framework, developer-tools, workflow-automation, testing
- domain: artificial-intelligence, developer-tools, tutorials
- platform: cli, cross-platform
- tags: ai-coding-agent, claude-code, autonomous-coding, prompt-engineering, experiment, nextjs, long-running-agents, ai-agents, automation, nodejs

## Member repositories
- SamuelQZQ/auto-coding-agent-demo (main) score 45

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:17.178107+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:07:50.919447+00:00, confidence not recorded.
  - readme: https://github.com/SamuelQZQ/auto-coding-agent-demo (fetched 2026-08-28T04:03:17.178107+00:00, sha b016e8d845ca)
- Data as of 2026-08-30T08:39:29.467469+00:00.