This piece summarizes a Jukebox Podcast interview with Damon Cook about WP Trend Watcher, an open source tool he built to help WordPress professionals keep up with the constant stream of news and updates.
Who is Damon Cook
Damon is a WordPress developer of more than a decade who recently moved from agency and hosting roles to running his freelance business, Noma Digital. He’s an engaged community member and contributor who likes to stay ahead of Core releases and ecosystem shifts so he can better serve clients and anticipate changes.
The problem he wanted to solve
WordPress produces a huge volume of content: core proposals, plugin and theme updates, blogs, community chatter, and social posts. Manually reading dozens of RSS feeds is time consuming and often inefficient—many items turn out to be irrelevant or duplicative. Damon wanted a way to surface the meaningful items quickly without losing the human judgment that prevents AI hallucinations.
What WP Trend Watcher does
WP Trend Watcher collects RSS feeds from a curated list of WordPress-related sources, runs the newest items through a local AI model, and produces concise summaries targeted toward developers and agency owners. The tool is designed to detect duplicates, filter weak or irrelevant items, and produce a readable weekly briefing with links.
Human in the loop
A core decision was to avoid fully automating publication. The AI creates initial summaries, but Damon reviews the output, clicks through original articles, and adds a human summary and commentary at the end of each report. This manual step serves two purposes: it validates the AI output to catch hallucinations or nuance the context, and it helps the information sink in for Damon so he retains what matters.
Technical setup
The codebase is TypeScript and is hosted on GitHub as an open source project. Damon runs the tool locally on his Mac, using LM Studio to load models that fit his hardware (he mentions working with models that fit around 18GB of RAM). The workflow is currently invoked manually—he runs it weekly, reviews and edits the draft, then publishes the report to GitHub Pages. Publication is automated after he commits the curated report, but the content creation remains deliberately human-curated.
Model experimentation and output
Damon has tested multiple local models and records which model produced each report. Outputs vary slightly by model, and he’s working on making final report formatting more deterministic and easier to footnote. The AI is used mainly as a curator and summarizer rather than as a creative author.
What gets filtered out
The pipeline includes checks for duplicate stories (same topic across different sites), weak language, and missing sections, so not every collected link becomes a published item. The AI is asked to favor factual, developer-relevant points rather than chasing popularity or clickbait.
Why he publishes it publicly
Damon is transparent about the process and the time he spends reviewing each report. Publishing the reports and the code invites others to fork, suggest sources, or improve the tool. He admits the process can feel ‘cheap’ compared to purely human curation, but transparency and the human-in-the-loop step help maintain trust.
Future possibilities
Damon envisions adding trend analysis that reviews past reports and draws longer-term inferences—why a topic rose or faded, what features evolved into Core, and other historical patterns. The architecture he’s building could be applied beyond WordPress to any domain with a corpus of RSS or feedable sources.
Tradeoffs and philosophy
The project highlights a common tradeoff with AI: automation can scale but human judgment remains vital, especially in technical communities where nuance matters. Damon intentionally resists full automation because the review step produces better, more trustworthy briefings and keeps him engaged with the material.
How to try it
The project is open source on Damon’s GitHub account (colorful-tones), and he documents setup steps on damonacook.com. WP Tavern’s episode page includes links to the repository and sample weekly reports so anyone can inspect the outputs and judge usefulness.
Bottom line
WP Trend Watcher is a practical, human-centered approach to using local AI as a curator for a noisy information space. It reduces the discovery burden by surfacing relevant items and saves time while preserving human verification and commentary. For freelancers, agency owners, and developers who want to stay informed without spending hours reading every source, it’s a useful scaffold and a promising foundation for future trend analysis.
