Hacker News Reader

link to documentation: https://docs.testchimp.io/test-planning/intro

nuwansam_87on User stories as docs in the repo instead of tickets

AI agents work best with plain .md files organized in folders.

Most work-planning tools today (Jira, Linear, etc.) are DB-centric and form-driven, which makes them awkward for agentic workflows.

I’ve been building an alternative where user stories and scenarios live as structured .md files syncable to your repo, with human workflows layered on top.

I’d appreciate feedback on where this fits (or doesn’t), and what you’d expect from an approach like this.

nuwansam_87on AI-wright: AI steps in Playwright scripts: open-source, Vision-enabled, BYOL

Here is a quick demo of it in action: https://youtu.be/MoMaXPnD5h8

nuwansam_87on Building AI Agents? Watch Memento (Or Ghajini for a Version with Song and Dance)

LLMs are deceivingly human like, which makes us often do the mistake of prompting them assuming such.

However, there is a key difference in how they perceive world state as agents - they live in snapshots unlike us. Putting your self in their shoes helps prompt them for significantly better results.