Damon Cook has been part of the WordPress ecosystem for over a decade — working at WordPress-focused agencies, contributing to the project, and recently running his freelance studio, Noma Digital. Frustrated by the volume of WordPress news flooding feeds, social networks, and developer channels, he built WP Trend Watcher: an open-source, local-first tool that sifts RSS sources, summarizes developments with AI, and keeps a human in the loop to verify and annotate the results.
Why build another news roundup?
For Damon it started selfishly: as a developer and freelancer he wants to stay one step ahead of what’s coming in core, plugins, and the broader ecosystem so he can anticipate client needs and understand shifting priorities. The WordPress landscape is vast and noisy — reliable sites appear and disappear, community conversations scatter across platforms — and manually scanning dozens of feeds every week is time-consuming and inefficient. WP Trend Watcher automates the discovery and initial summarization of items, leaving the critical human judgment to confirm, expand, and contextualize what matters.
How WP Trend Watcher works
At a high level the tool collects RSS feeds, deduplicates overlapping coverage, and runs article content through a prompt-driven summarization step. Damon prefers running models locally: the project is a TypeScript codebase you can fork from GitHub, and he runs inference using models installed on his Mac (via tools such as LM Studio). The output is a concise, numbered weekly briefing consisting of headlines, short summaries tailored for developers and agencies, and links to source articles.
Crucially, WP Trend Watcher is not fully automated. Damon intentionally enforces a “human in the loop” stage: after the AI produces summaries he reviews the items, clicks through to verify accuracy, and adds freeform human notes at the end of each report. That step serves two purposes: it catches hallucinations or weak inferences from the model, and it helps the information register in his own memory so the trends actually stick.
Technical details and workflow
– Codebase: TypeScript, available on GitHub for forking and contribution. – Scheduling: Damon runs the report weekly (a configurable choice) and the tool grabs new items since the last run. – Models: He experiments with various local LLMs that fit his hardware (choices and model names vary week to week). – Deduplication: The summarizer checks for duplicate coverage and filters out redundant items. – Publishing: After Damon reviews and annotates a report, he publishes it using GitHub Pages. The human review step is what prevents full automation.
The advantage of local models is privacy and experimentation: Damon swaps models to compare results and records which model produced each report. He also aims to make the final output more deterministic and include clearer footnotes and source references in future iterations.
Why the human element matters
AI can quickly surface likely-relevant headlines from a large corpus, but it can also invent details or miss nuance. Damon treats the summarization phase as a triage step — the AI points to potential items worth reading, and he performs the verification. He is transparent about the process: each report lists how much time he spent reviewing, so readers can decide how much confidence to place in it. That transparency also makes the project useful for others: you can fork it, contribute sources, or change the filtering criteria.
What’s left out and how sources are chosen
The tool tries to avoid repetition and strips out superfluous or weakly supported items. It prioritizes factual, developer-centric pieces over “trendiness” for its own sake — the name WP Trend Watcher refers more to tracking substantive developments than chasing popularity metrics. Damon also provides a way for others to suggest additional RSS sources, but he warns of scope creep: more feeds means more noise, and part of the challenge is keeping the feed set focused and useful.
Use cases and future potential
For freelancers, agency leads, and plugin/theme authors the briefing acts as a time-saver and early-warning system: instead of spending hours hunting for updates, you get a curated list of items to inspect. Damon envisions additional layers over time: aggregating past reports to detect genuine trends, asking the system to explain why a feature appeared in core or how a market metric shifted, and applying the same approach to other domains beyond WordPress.
Because the project is open source, it can be repurposed for any niche where many sources produce continuous updates — from local clubs to entire industries. Keeping “receipts” in sequential reports also enables retrospective analysis: years from now you could examine how a feature made its way into core by studying the chain of discussions and announcements.
Where to find it
Damon publishes the weekly briefs and maintains the repository on GitHub; he also documents setup steps on his site. The project uses local AI models and requires a modest hardware footprint to run locally, though models are continually shrinking and becoming more accessible.
Final thoughts
WP Trend Watcher is a practical, transparent approach to making AI useful for busy WordPress professionals. It accepts automation’s strengths (fast triage and summarization) while explicitly preserving human judgment where it matters. For those overwhelmed by the volume of WordPress news, it’s a strong example of how local models, RSS, and a human reviewer can combine into a reliable weekly briefing — and a scaffold for richer trend analysis down the road.
