agentsclimarketplace

Weekly lit review

Skill zhongguojie1998/week-lit-review/skills/weekly-lit-review

Run the genomics literature review pipeline. Searches bioRxiv and 15 journal RSS feeds, downloads PDFs, then Claude reads each PDF and writes critical reviews with scores. Supports an interactive daily mode (--interactive): query titles/abstracts, pick papers from a selectable list, then deep-review only the chosen ones. No API key needed — works entirely within Claude Code. Triggers: "literature review", "weekly papers", "journal scan", "paper review", "genomics review", "preprint screening", "lit review", "daily papers".From its SKILL.md

Install
npx -y skills add zhongguojie1998/week-lit-review --skill weekly-lit-review

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.9 KB, as published. Nobody here has run it

Weekly Genomics Literature Review

A literature-review pipeline. To keep context small, the detailed steps live in per-mode reference files — read only the one that matches the invocation, plus the shared review/score procedure.

Mode Detection

Pick exactly one mode from $ARGUMENTS, then read that file and follow it:

If arguments contain…Read and follow
--interactive${CLAUDE_PLUGIN_ROOT}/skills/weekly-lit-review/references/mode-interactive.md
--gmail-stork${CLAUDE_PLUGIN_ROOT}/skills/weekly-lit-review/references/mode-gmail-stork.md
one or more --doi${CLAUDE_PLUGIN_ROOT}/skills/weekly-lit-review/references/mode-doi.md
otherwise (default)${CLAUDE_PLUGIN_ROOT}/skills/weekly-lit-review/references/mode-batch.md

Every mode reuses the shared procedure in ${CLAUDE_PLUGIN_ROOT}/skills/weekly-lit-review/references/review-and-score.md (how to read a paper, the 7 review sections, 0–10 scoring, and how reviews/summary HTML are produced via the render scripts). Read it when you reach the review step.

Common setup (all modes)

  1. Ensure directories exist:

    mkdir -p ~/Desktop/Claude/week-lit-review-results/{source,reviews,$(date +%Y-%m-%d)}
    

    Layout: source/ (shared downloaded files), reviews/ (shared review HTML + JSON archive — the canonical store for dedup), {YYYY-MM-DD}/ (per-run manifest, logs, _index.json, and summary.html).

  2. Read the config for search parameters (biorxiv_categories, genomics_keywords, journal_feeds, review_engine):

    Read: ${CLAUDE_PLUGIN_ROOT}/assets/config.yaml
    

{TODAY} below means $(date +%Y-%m-%d).

Output conventions

  • Reviews and summary are written by render scripts (scripts/render_review.py, scripts/render_summary.py) — do not write HTML by hand. Per-paper reviews come from compact JSON; the summary is built in pure Python from the persisted review JSONs (no hand-written summary JSON). Details and schemas are in references/review-and-score.md.
  • Individual reviews (HTML + JSON): ~/Desktop/Claude/week-lit-review-results/reviews/{filename}.{html,json}
  • Per-run summary: ~/Desktop/Claude/week-lit-review-results/{TODAY}/summary.html

Keep looking

Skills are one crate of 608,475. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.