Verified:GitHub

lumi-research-watch-run

Run scheduled watchlist discovery and return a JSON summary of new paper candidates.

36|16|Updated May 1, 2026

Search Tags:#research #watchlist #paper discovery #cron automation #rss discovery #scheduled polling #stateful ingestion

Ready to run

npx skills add https://github.com/tronghieu/lumina-wiki --skill lumi-research-watch-run

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: lumi-research-watch-run
Source: https://github.com/tronghieu/lumina-wiki/tree/main/src/skills/packs/research/watch-run
Command: npx skills add https://github.com/tronghieu/lumina-wiki --skill lumi-research-watch-run

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

It helps you quickly discover newly available papers and feed items from your consolidated watchlist without manually checking sources one by one.

Core Features & Use Cases

Scheduled discovery runner: Triggers a single, manual discovery pass over your watchlist (topics and feeds) when you want to “check for new papers” or “run discovery now”.
Stateful per-feed polling: Updates per-feed state files under _lumina/_state/feeds/ and writes discovered candidates to raw/discovered/ while leaving wiki/ untouched.
Readable results summary: Produces a plain-language breakdown of how many new candidates (or errors) appeared per watchlist item, with actionable next steps.

Quick Start

Use the lumi-research-watch-run skill to run one discovery pass over your watchlist and get a plain-language summary of new candidates.

Frequently Asked Questions about lumi-research-watch-run

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I run watchlist discovery to find new research papers automatically?▼

You run watchlist discovery by triggering a one-shot discovery pass against your watchlist configuration file. This process polls your configured topics and feeds, writes new paper candidates to a raw discovered directory, and returns a JSON summary of the findings.

Can I use cron automation for scheduled polling of RSS feeds?▼

Yes, cron automation supports scheduled polling of RSS feeds for new papers. The discovery runner integrates with cron, launchd, or Task Scheduler to trigger recurring discovery passes, automatically checking your watchlist for new items at set intervals.

What is stateful ingestion and how does it work for paper discovery?▼

Stateful ingestion tracks the polling status of each RSS feed by saving state files under a dedicated feeds directory. This mechanism ensures the discovery runner remembers previous checks, preventing duplicate paper candidates from appearing in subsequent discovery passes.

Does watchlist discovery modify my existing wiki or notes?▼

No, watchlist discovery does not modify your wiki. The discovery runner explicitly leaves wiki directories untouched, only writing newly discovered paper candidates to a raw discovered folder and updating internal feed state files.

What do I need to configure before checking for new papers?▼

Before checking for new papers, you need a watchlist configuration file defining your desired topics and RSS feeds. The discovery runner requires this YAML configuration file to execute a discovery pass and identify relevant new paper candidates.