Tech

Make Less Data: Self-Host Miniflux Feed Reader

This is the first in a series about self-hosted services that you can run yourself in order for you to be less of a product of other companies. This means you will be creating less marketable data for these companies and likely saving money in the process.

The shutdown of Google Reader years ago sent ripples across the internet as there was now room for competition in the form of small, paid feed reading services like Feedbin. Having been a customer of Feedbin for several years and being happy with it, I nevertheless decided to explore options for self hosting an RSS reader service as an exercise of being more self sufficient. It is even more incentive that as more apps and services become subscription-based, I wanted to reduce my growing number of recurring software subscriptions as much as possible.

Agentic Coding Demands a Higher Code Quality Baseline

The age of agentic coding requires a greater focus on the baseline of code quality.

In the past, great developers often had a mindset of leaving something better than they found it. In the slack time between ticket estimates, engineers would often dedicate time to refactor something that had grown stale, alongside pushing new development forward.

Today, code is generated quickly and cheaply, with more focus given to pushing new development forward at a blistering pace. This leaves less time to evaluate the quality of existing code for opportunities to improve it. I have found that the biggest impact on agentic code quality is the existing code quality, with AGENTS.md, prompts, and skills having much less of an impact than what is already in place.

Save YouTube Video Learnings to Obsidian

I use the note-taking app Obsidian for my personal organization needs, and over time I’ve been building it to function as a second brain. Since Obsidian is built around markdown files on the file system, there’s no vendor lock-in, and it’s very easy to use with AI models for reading and writing to it. This has led me to optimize how I retain information that I want to remember.

Occasionally, I’ll watch an inspiring YouTube video that teaches me something useful. Bookmarking the link is futile because no knowledge is retained, and I would almost never go back into my bookmarks and watch it again. The experience was mostly ephemeral. Not wanting interesting content to be transitory, I set out to build a skill that distills the learning from the video’s audio into an Obsidian note that’s easy to re-read in the future. The cleaned-up note becomes part of my knowledge base to inspire me and help me learn from the content.

Claude Code Skills for Obsidian Personal Knowledge Management

Taking notes somewhere is a great way to augment your brain’s memory with an external storage location. The term Personal Knowledge Management (PKM)[1] is becoming more common to describe the new era beyond flat notes that require digging up now that AI is possible to combine with one of these note-taking systems. It is now possible to store information that can be recalled, parsed, and reasoned about, versus in the past when note-taking was more input than output due to the effort to search and dig up old notes. Having a virtual brain connected to your personal notes creates a new era of memory and learning. To manage this, I have created a set of agent skills for Claude that create a consistent workflow for using Claude Code with my notes effectively.