How to use Claude to maintain public data sites
An author reports using Claude for a public water data site and expanding to nine more sites, with automated weekly traffic analysis and article updates. Backlink outreach remained a planned human task.
news.ycombinator.com
Prompt to try
Editorial adaptation, not the source author’s exact prompt. Replace bracketed inputs and review tool permissions first.
Build [public data site] using [government data source]. Update the site and prepare [newsletter]. Review [GA4 data] and [Search Console queries] weekly, then propose or make article updates that address observed demand. Leave outreach emails for me.
Reach people exploring AI agents
See placements →What happened
The Waterline began receiving search impressions and clicks. The author launched nine more sites and reported The Dwell was doing best for traffic, with weekly content improvement automated.
Reported by idopmstuff on Hacker NewsHow the agent did it
- Launch a public water data website with Claude.
- Use ChatGPT site ideas to launch nine additional sites.
- Request weekly analysis of GA4 and Search Console data.
Where a person stepped in
The post does not describe content review steps. The author explicitly kept prospective backlink emails under human control.
A person set the goal and reviewed the result.
The author launched The Waterline as a side project and an evaluation of whether Claude could handle a substantial project mostly independently. The site pulls public government data about water levels in the western United States, updates its website, and sends a newsletter. After it began receiving search impressions and clicks, the author expanded the approach to additional sites.
ChatGPT supplied ten ideas for this type of site. The author reports launching the remaining nine, with The Dwell, covering ship and truck dwell times and port congestion, performing best for traffic at the time of the post. Initially, the author had the system send a weekly analysis of GA4 and Google Search Console data to identify improvements. That process was subsequently automated: each week it reviews traffic and search queries, then updates existing articles or creates new ones to address demand.
The author describes this as, in theory, an engine for launching sites that improve themselves over time. Backlinks remain a need. ChatGPT generated possible linking prospects, but the author explicitly declined to let an agent email them and planned personal outreach instead.
Who it's for: People maintaining public data publications who can define their data sources and retain responsibility for outreach.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · idopmstuff
Frequently asked questions
What data does The Waterline publish?
The author says it pulls public government data about water levels in the western United States, updates the website, and sends a newsletter.
How does the weekly improvement process work?
The system reviews traffic and search queries each week, then updates articles or creates new ones to meet demand. The author names GA4 and Google Search Console as the analysis inputs.
Did the author automate backlink outreach?
No. ChatGPT generated lists of possible linking prospects, but the author was unwilling to let an AI agent email people and planned to send selected emails personally.
More details
Remaining steps
- Automate weekly traffic reviews and article updates.
About this account
Claude handled the initial project mostly independently; the later system reviewed traffic and revised or created articles weekly.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=49235288. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: idopmstuff on Hacker News
Source published 9 Aug 2026
Source checked 14 Oct 2026
Tools and topics
Claude · ChatGPT · GA4 · Google Search Console
Public data · SEO · Analytics
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.