How to use Claude to build an EPUB toolkit
A nondeveloper used Claude Code to expand a personal EPUB merger into an online toolkit with nine tools, a Pro plan, REST API and MCP server.
news.ycombinator.com
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Help me build [EPUB utility] on Cloudflare Workers with Hono and vanilla JavaScript. Start with [required operation], explain in-memory processing constraints, and help me verify the result before adding further tools.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source poster / publisher
- Seventeen18
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The author published epublys.com with nine tools and says it covers their needs. They report low running costs and acknowledge potentially unfinished security work.
- Human involvement
- The author directed feature expansion and pricing and said their needs were covered; the post does not describe formal review or testing.
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
- Start building an EPUB merger with Claude Code on Cloudflare Workers.
- Expand the utility into nine EPUB and document tools.
- Put the service online after assessing its running costs.
The author wanted to merge EPUB files without opening Calibre or using online services that required accounts or displayed ads. They started building a small tool on Cloudflare Workers with Claude Code. That initial merger grew into nine tools: merging, splitting, compression, EPUB-to-PDF conversion, PDF-to-EPUB conversion, metadata editing, validation, a reader and an AI cover generator using FLUX.
The reported stack uses Hono on Workers, vanilla JavaScript on the frontend, and jszip and pdf-lib for file processing. Files are processed in memory because Workers do not provide a filesystem. The author identifies the memory and CPU limits as constraints that shaped the implementation.
After finding operating costs low, they put the service online. They added a $9 monthly Pro plan offering larger files and branding removal, plus a REST API and MCP server. The author says their own needs are covered, while acknowledging that more security work might remain. They provide no user adoption figures.
Who it's for: People building a focused file utility around their own recurring needs.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · Seventeen18
Frequently asked questions
Which EPUB tools did the author build?
They list merging, splitting, compression, EPUB-to-PDF and PDF-to-EPUB conversion, metadata editing, validation, a reader and a FLUX-based AI cover generator.
What did the paid plan offer?
The reported $9 monthly Pro plan supported 200MB files rather than 10MB files, or removal of branding.
More details
Remaining steps
- Add a Pro plan, REST API and MCP server.
About this account
Claude Code assisted development of the toolkit; the source does not specify which implementation decisions or actions it performed independently.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=47586230. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: Seventeen18 on Hacker News
Source published
Added to Agents In Use
Source checked
Tools and topics
Claude Code · Cloudflare Workers · Hono · jszip · pdf-lib · FLUX
EPUB · Cloudflare Workers · Personal software
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.