News Monitor

Inspired by USV's News Monitor, rebuilt as a generic version for any team that keeps a folder of who and what it tracks. It searches each tracked name on a short list of named sites (TechCrunch, The Information, Ars Technica by default, and you can edit the list) — or, if you'd rather skip live search, you can hand it a news export instead. Every item is checked against your own files first, never guessed. An item that doesn't name anyone you track gets dropped, not reported. An item that could belong to two different people or companies is kept under both, never guessed onto one. Every kept item carries a quote so you can see why it matched. It keeps at most 3 stories per name, so the digest stays short, and an entity with nothing found gets an honest zero-result line instead of padding. It only ever reads your files — it never edits a person, organization, or meeting file, and only writes to its own digests folder.

First-party, by Skills and Agents Co

August 22, 2026

Install

Simple install — no setup required

1

Open the Claude desktop app and go to the Project you want this in

2

Go to Code, then Customize, then click Create a new skill and upload the skill file below

3

Say a trigger phrase to activate it

The skill file
Download
Or install with the CLInpx skills add skills-agents-co/skills-and-agents-library/news-monitor

What this skill does

This is an overview of the skill from the site, not the skill's real instructions. The full SKILL.md Claude reads when this skill is active lives on GitHub at the source below.

Inspired by USV’s News Monitor (Union Square Ventures) — USV built theirs to surface news on the companies and people in their portfolio without wading through a firehose. This is a generic, independently built version, not their code: any team that keeps files on people, organizations, or projects can point this at that same folder and get the same shape of output.

Point the skill at the entity folder meeting-scribe writes and calendar-agent reads, and it either reads a news export you hand it (RSS/Atom export, saved search page, forwarded newsletter) or searches each tracked name on a fixed default list of sites. Every item is checked against your entity files first instead of guessed from search-result text, and written into one dated digest. A matched item shows under the matching entity with a grounding quote. An item naming no tracked entity is dropped entirely — it never shows up, not even as unmatched. An item that could be two different tracked entities gets flagged as ambiguous, listed under both, again with nothing guessed. An entity with nothing relevant found this run gets a plain “no relevant news found” line. It only ever writes to its own digests folder — it won’t touch your people, company, or meeting files. Run calendar-agent before a meeting to prep, meeting-scribe after it to record, and news-monitor on its own schedule to watch for news on the same tracked list in between.

View full SKILL.md on GitHub