How to use Claude to map territories from legal text
Claude helped a workplace user assemble scattered industry knowledge, generate territory polygons from legal descriptions, and create a frontend. The author reports sufficient precision in checks completed so far.
news.ycombinator.com
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Using [industry source materials], assemble a dataset for [internal use case]. Convert the written territory definitions in [legal documents] into geolocation polygons, flag ambiguities for review, and build a frontend for [required tasks]. Target [precision requirement] and provide outputs for human checking.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source poster / publisher
- zhoBEENG
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The author reports a highly usable frontend and a large dataset produced over a weekend, with negligible errors and sufficient polygon precision in checks completed so far.
- Human involvement
- Humans participated and checked some generated data, but the post does not specify their roles or validation method.
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
- Use Claude to assemble a dataset of scattered industry-specific knowledge.
- Generate geolocation polygons from legal documents describing physical territories.
- Use Claude to create a frontend backed by the dataset.
A user described using Claude over a weekend to build a small project for a nontechnical workplace. They deliberately withheld industry and project details, but identified two main outputs: a highly usable frontend and a large dataset of industry-specific knowledge that was not available from any single public source.
Part of assembling that dataset involved generating geolocation polygons from legal documents that described physical territories in words. The author says absolute precision was unnecessary for their purpose. Based on the checks completed so far, they reported that the agent produced the data largely in one attempt, with negligible errors and sufficient precision.
Humans participated in the process, although the post does not explain their exact responsibilities or the validation method. The author said the dataset alone would have taken weeks of full-time effort and that the project would not have happened without agents. Their claim of significant workplace value is prospective: the post describes the built dataset and frontend, but gives no measured operational impact or deployment details.
Who it's for: Teams exploring an internal tool that combines scattered domain knowledge with approximate geographic representations of written territory definitions.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · zhoBEENG
Frequently asked questions
What did Claude produce for the workplace project?
The author reports a usable frontend and a large industry-specific dataset, including geolocation polygons generated from written territory definitions in legal documents.
How precise were the generated territory polygons?
The author says absolute precision was unnecessary and that checks completed so far found negligible errors and sufficient precision. The post does not describe the checking method.
More details
Remaining steps
- Check the generated data for errors and sufficient precision.
About this account
Claude helped generate the dataset, territory polygons, and frontend; the source does not detail execution permissions or the interaction pattern.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=48529183. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: zhoBEENG on Hacker News
Source published
Added to Agents In Use
Source checked
Tools and topics
Claude
Geolocation · Legal documents · Datasets · Internal tools
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.