skills.rest
yakeworld
yakeworldCOMMUNITY
Explore all skills by this author →
GitHub

research

Monitors PubMed and arXiv literature across multiple research directions with local date filtering.

6|1|Updated May 11, 2026
Search Tags:#research#api#deduplication#fallback#pubmed#arxiv#literature monitoring
One-click install
npx skills add https://github.com/yakeworld/Synthos --skill research-yakeworld
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/yakeworld/Synthos/tree/main/skills/private/research
Command: npx skills add https://github.com/yakeworld/Synthos --skill research-yakeworld

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Keeping up with new publications across several research directions requires repetitive manual searches on PubMed and arXiv, and naive API queries return noisy, poorly dated, or duplicated results that waste screening time. ## Core Features & Use Cases - Multi-Direction Literature Monitoring: Define 2-5 keyword variants per direction and query PubMed E-utilities and arXiv API in parallel for maximum recall. - Local Date Filtering & Deduplication: Fetch wide-range results, filter locally by PubMed Year or arXiv published fields (avoiding the unreliable reldate parameter), deduplicate by exact title match, and rank by keyword match plus recency weighting. - Fallback Strategies: When standard tools are blocked or unavailable, fall back to direct curl calls, curl-to-file plus standalone python3 processing, or RSS substitutes; rewrite noisy arXiv queries with precise Boolean expressions like all:eye+AND+all:tracking. - Use Case: Track pupil tracking and gaze analysis literature over the last 3 months, producing a deduplicated Top-N list per direction with relevance scores. ## Quick Start Ask the agent to monitor recent publications on pupil tracking and gaze estimation from PubMed and arXiv over the last three months and return a deduplicated ranked Top 10 list per direction.

Frequently Asked Questions about research

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

FAQPage Schema
How do I monitor new PubMed and arXiv papers for multiple research topics?▼

Define 2-5 keyword variants per direction, then query PubMed E-utilities (esearch.fcgi plus efetch.fcgi) and the arXiv API in parallel. Filter results locally by date, deduplicate by exact title match, and rank by keyword match plus recency weighting before selecting a Top-N list.

Why should I avoid the PubMed reldate parameter for date filtering?▼

The reldate endpoint parameter is unreliable for precise time windows. Fetch a wide result range from the API, then parse the PubMed Year field or arXiv published field locally and filter to your target window, such as the last three months.

How do I reduce noise in arXiv search results like false 'eye tracking' matches?▼

Broad queries such as 'eye tracking' match unrelated papers on particle tracking or counterfactual tracking. Use precise Boolean expressions like all:eye+AND+all:tracking to restrict matches, and manually screen the ranked output before finalizing the Top-N list.

What can I do when curl or inline scripts are blocked by a security scan?▼

Use the documented fallback path: curl the API response to a file first, then process it with a separate standalone python3 script. Other fallbacks include calling the arXiv or PubMed APIs directly when SearXNG is down, or using RSS feeds when the HN Firebase API is slow.

Does this literature monitoring workflow depend on HuggingFace or SearXNG?▼

No. HuggingFace may be unreachable from some servers and must not be a required step, and SearXNG has a direct-API fallback. The core retrieval path relies only on the PubMed E-utilities and arXiv APIs.

Related Skills

Explore More Education & Research Skills→
MSiccDevMSiccDev
20

arc42-section-12

Scan arc42 sections to generate an alphabetized glossary with definitions and aliases.

Official
Advanced
mobled37mobled37

research-pipeline

Automate academic research pipelines from literature review to experiment planning.

Community
Advanced
Architekt-JutraArchitekt-Jutra
40

research-gatherer

Collect and cross-verify raw research findings from internal and external sources.

Official
Advanced
xjtulycxjtulyc
33

systematic-review

Transform PICO inputs into PRISMA 2020 systematic review workflows.

Community
Advanced