← ClaudeAtlas

bioresearcher-pubmed-weeklylisted

Downloads the past week's PubMed daily update XML.gz files from ftp.ncbi.nlm.nih.gov/pubmed/updatefiles/, streams-parses interleaved PubmedArticle and DeleteCitation records, and produces one combined Excel workbook plus a summary JSON. Use for weekly PubMed/NLM literature tracking, NCBI updatefiles retrieval, pubmedNNnNNNN.xml.gz parsing, PMID/DOI/journal/author extraction, deleted-PMID lists, or building a weekly biomedical literature table.
yeyuan98/bioresearcher-skills · ★ 1 · Data & Documents · score 73
Install: claude install-skill yeyuan98/bioresearcher-skills
# PubMed Weekly Update Download and Parse This skill downloads the past week's (Monday-Sunday) PubMed daily update `xml.gz` files from `ftp://ftp.ncbi.nlm.nih.gov/pubmed/updatefiles/`, parses them with the bundled streaming scripts, and produces ONE combined Excel workbook (`combined.xlsx`) plus a summary JSON. ## Workflow Overview 1. **Python environment check**: ensure uv is available (install via the `bioresearcher-python-setup-uv` skill if needed) 2. **Date-range calculation**: compute the past week's Monday-Sunday range 3. **FTP listing + filtering**: fetch available `xml.gz` updatefiles and keep those modified within the week 4. **Download with retry + resume**: download each filtered file (3 attempts, 2s delay, `.part` resume) 5. **Parse + combine via scripts**: parse all downloaded files and produce one Excel workbook 6. **Report**: verify outputs and summarize to the user ## Prerequisites - Internet connection and access to the NCBI FTP server - uv package manager (if missing, load the `bioresearcher-python-setup-uv` skill and follow it EXACTLY, then return here) - openpyxl is provided at runtime via `uv run --with openpyxl` ## Script Usage Replace `<skill_dir>` with the full path to this skill's directory (the scripts live in `<skill_dir>/scripts/`). Run all commands from the working directory where downloads should land. **For Unix-like shells (Git Bash / macOS / Linux):** ```bash uv run --with openpyxl python <skill_dir>/scripts/pubmed_weekly.py <command>