Using Claude Code to build a browser game-asset manager
@dsmart had been tracking assets for two games in a spreadsheet. He reports that his Claude Code Opus agents built a fully functional browser-based asset manager in exactly 23 minutes, with no fixes or changes needed from him.
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Build a browser-based asset manager for my game projects using the requirements I provide. Explain the implementation, show how to run it locally and list what still needs testing.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source poster / publisher
- @dsmart
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- @dsmart reports a fully functional browser-based asset manager built by his Claude Code Opus agents in exactly 23 minutes, which he timed. He says he needed no fixes or changes.
- Human involvement
- The poster identified the asset-management problem and timed the build. He reports making no corrections; prompts, review methods and deployment approvals are not supplied.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- @dsmart had been using a spreadsheet to track assets while working on two games.
- His Claude Code Opus agents built a browser-based asset manager.
- He timed the build at exactly 23 minutes and describes the result as fully functional.
@dsmart was working on two games and tracking their assets in a spreadsheet. He says existing digital asset managers were bloated, ineffective or tied to particular game engines. That dissatisfaction provided the reason to build a browser-based manager suited to his own work.
He reports that his Claude Code Opus agents built a fully functional version in exactly 23 minutes, adding that he timed it. He also says he did not have to fix or change anything because it worked. Both the duration and the lack of corrections are his reported experience, not a benchmark or an independently tested result.
The post establishes the spreadsheet as the previous tracking method. It does not say the agents imported the spreadsheet, describe the prompt, identify a framework or list the application’s features. There is no published repository or demonstration in the supplied capture, and no media is attached according to Scout’s API lookup.
The supported use case is a custom browser application built by coding agents for a game-development task. The account does not establish production reliability, security, deployment arrangements or how many agents were involved. Human direction is evident in choosing the problem; implementation and testing checkpoints are otherwise unspecified.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: X · @dsmart
Frequently asked questions
Was the spreadsheet imported into the new app?
The post says a spreadsheet was the previous asset tracker. It does not state that agents imported it or used it as a specification.
How was the 23-minute figure measured?
The poster explicitly says he timed it and reports exactly 23 minutes. The capture contains no independent timing record.
What can be verified about the finished application?
The source calls it fully functional and says no corrections were needed, but supplies no code, feature list or independent test results.
More details
Remaining steps
- He reports needing no fixes or changes; the post does not describe testing or deployment steps.
About this account
The poster attributes implementation of the browser-based asset manager to Claude Code Opus agents. The division of work, tool calls and validation process are not described.
Based on Scout’s verbatim X public API capture, read live on 2026-11-01 and supplied with explicit permission instead of opening x.com. Capture text, publication context and supplied author bio were reviewed; no maker affiliation is stated in that bio. No self-thread continuation or media content was reviewed. Outcomes are self-reported, not independently reproduced. Suggested prompt is an editorial addition; no exact operational task prompt is established by this capture.
Credit: @dsmart
Source published
Added to Agents In Use
Source checked
Tools and topics
Claude Code
Game assets · Browser application · Custom software
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.