skills.rest
MathematicGuy
MathematicGuyCOMMUNITY
Explore all skills by this author →
GitHub

lumi-research-watchlist

Configure scheduled research topic watchlists in YAML for automated paper discovery.

Updated Jun 3, 2026
Search Tags:#research#configuration#yaml#arxiv#watchlist#literature monitoring#scheduled discovery
One-click install
npx skills add https://github.com/MathematicGuy/team-006-vin20k --skill lumi-research-watchlist-mathematicguy
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: lumi-research-watchlist
Source: https://github.com/MathematicGuy/team-006-vin20k/tree/main/lumina-wiki/.agents/skills/lumi-research-watchlist
Command: npx skills add https://github.com/MathematicGuy/team-006-vin20k --skill lumi-research-watchlist-mathematicguy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Managing which research topics to track over time usually means hand-editing configuration files and remembering scheduler syntax. This Skill turns plain-language requests into validated watchlist entries so users can follow research topics without touching YAML structure directly. ## Core Features & Use Cases - Topic Management: Add, change, pause, resume, or remove research topics in _lumina/config/watchlist.yml using natural language. - Source and Schedule Mapping: Converts user intent into concrete fields like sources (arxiv, s2, openalex), schedule (manual, daily, weekly, monthly), and result limits. - Validation: Runs a dry-run check via lumina discover run --dry-run --json or the discover-runner script, with manual YAML shape inspection as fallback. - Use Case: A researcher says "track LLM agent memory papers weekly" and the Skill appends a properly formatted, validated watchlist item without touching any other project files. ## Quick Start Ask the assistant to add a weekly watchlist topic for a research phrase you want to follow, and confirm the generated entry.

Frequently Asked Questions about lumi-research-watchlist

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

FAQPage Schema
How do I add a research topic to a scheduled discovery watchlist?▼

Describe the topic in plain language, including how often to check and which sources to use. The Skill appends a YAML item with id, query, sources, schedule, limit, and max_new fields to the watchlist file, then validates it with a dry run.

What sources can a research watchlist query for new papers?▼

Supported sources are arxiv by default, Semantic Scholar (s2) when configured, and OpenAlex for broader coverage including humanities and biomedicine. OpenAlex works without credentials for small checks, with an optional API key for daily budget tracking.

Can I pause a watchlist topic without deleting it?▼

Yes, pausing only flips the enabled field to false, preserving the topic configuration for later resumption. The Skill recommends pausing over removal since deleting a topic does not remove already discovered candidates anyway.

Does the watchlist skill run the discovery or download papers?▼

No, it only edits and validates the watchlist configuration. Scheduled discovery requires an external scheduler like cron or GitHub Actions, and PDF downloads happen later during a separate ingest step after you choose candidates.

Why does watchlist validation fail after editing the YAML?▼

Common causes are duplicate or non-lowercase id values, invalid schedule entries outside manual/daily/weekly/monthly, or missing version and items fields. The Skill reports these in plain language and fixes them before finishing.

Related Skills

Explore More Education & Research Skills→
infinitywingsinfinitywings
13

mcp-credentials

Automate credential setup and validation for MCP servers.

Community
Intermediate
tronghieutronghieu
36

lumi-research-watch-run

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

Community
Intermediate
JansenAnalyticsJansenAnalytics
5

rss-digest

Automate RSS/Atom feed subscriptions, keyword filters, and Telegram digests.

Community
Intermediate
Yuan1z0825Yuan1z0825
38.3k

nature-literature-pipeline

Automates multi-source literature discovery, scoring, digest delivery, and archival via scheduled pipelines.

Community
Advanced