citation-management
npx machina-cli add skill arendon1/agent-skills/citation-management --openclawCitation Management
Overview
Manage citations systematically throughout the research and writing process. This skill provides tools for searching academic databases (Google Scholar, PubMed), extracting accurate metadata, validating citation information, and generating properly formatted BibTeX entries.
Core Capabilities:
- Search: Find papers on Google Scholar and PubMed.
- Extract: Convert DOIs/PMIDs/URLs to complete BibTeX.
- Format: Clean and standardize BibTeX files.
- Validate: Check for errors, broken DOIs, and duplicates.
Core Workflow
Phase 1: Search
Detailed Guides:
- Google Scholar Search:
scripts/search_google_scholar.py - PubMed Search:
scripts/search_pubmed.py
# Google Scholar
python scripts/search_google_scholar.py "topic" --limit 50 --output results.json
# PubMed with MeSH
python scripts/search_pubmed.py --query '"Topic"[MeSH]' --output results.json
Phase 2: Metadata Extraction
Detailed Guide: Metadata Extraction
# From DOI/PMID/ArXiv/URL
python scripts/extract_metadata.py --doi 10.1038/nature12345
python scripts/extract_metadata.py --pmid 12345678
python scripts/extract_metadata.py --arxiv 2301.00000
# Batch from file
python scripts/extract_metadata.py --input ids.txt --output refs.bib
Phase 3: BibTeX Formatting
Detailed Guide: BibTeX Formatting
# Clean, Sort, Deduplicate
python scripts/format_bibtex.py raw.bib \
--deduplicate \
--sort year \
--output clean.bib
Phase 4: Validation
Detailed Guide: Citation Validation
# Check accuracy and fix common errors
python scripts/validate_citations.py clean.bib \
--auto-fix \
--report report.json \
--output final.bib
Quick Reference: Scripts and Tools
| Script | Purpose |
|---|---|
search_google_scholar.py | Search Google Scholar for papers. |
search_pubmed.py | Search PubMed using E-utilities. |
extract_metadata.py | Get metadata from identifiers (DOI, PMID, etc.). |
doi_to_bibtex.py | Fast DOI-to-BibTeX conversion. |
format_bibtex.py | Clean, sort, and deduplicate BibTeX files. |
validate_citations.py | Verify accuracy and find errors. |
Integration
This skill is designed to be orchestrated by the Academic Author skill.
- Academic Author: Writing and structure.
- Citation Management: Reference accuracy and formatting.
- Scientific Schematics: Diagram generation.
See ../academic-author/SKILL.md for the overarching workflow.
Source
git clone https://github.com/arendon1/agent-skills/blob/master/academic-author/subskills/citation-management/SKILL.mdView on GitHub Overview
Manage citations systematically across the research lifecycle. This skill searches Google Scholar and PubMed, extracts metadata, and converts it into clean BibTeX entries while validating accuracy and preventing duplicates.
How This Skill Works
The workflow follows four phases: search, metadata extraction, BibTeX formatting, and validation. It leverages scripts like search_google_scholar.py, search_pubmed.py, extract_metadata.py, format_bibtex.py, and validate_citations.py to build a final, clean BibTeX file.
When to Use It
- Building a literature foundation by collecting papers on a topic from Google Scholar and PubMed.
- Preparing a manuscript's bibliography with clean, standardized BibTeX entries.
- Verifying DOIs, PMIDs, and URLs to ensure citation accuracy and completeness.
- Batch-processing a list of identifiers from a file to generate a BibTeX library.
- Integrating citation workflows within the Academic Author skill for end-to-end research tasks.
Quick Start
- Step 1: Search Google Scholar and PubMed for your topic with the provided scripts.
- Step 2: Extract metadata from DOIs/PMIDs/URLs to generate BibTeX entries.
- Step 3: Format, deduplicate, and validate to produce final.bib (finalized BibTeX).
Best Practices
- Always run the validation step to catch broken DOIs and duplicates.
- Deduplicate and sort BibTeX by year to improve readability.
- Batch-process identifiers from a file to stay consistent across projects.
- Cross-check metadata against primary sources when possible.
- Version-control the final BibTeX file alongside your manuscript.
Example Use Cases
- Researcher builds a topic-specific BibTeX library by searching Google Scholar and PubMed, then validates and formats for a manuscript.
- Student converts DOIs to BibTeX, cleans entries, deduplicates, and sorts by year for a thesis bibliography.
- Team creates a shared reference file by batch-extracting metadata from multiple IDs and exporting clean.bib.
- Lab updates citations after new papers publish by re-running the search and re-validating entries.
- Grant proposal requires fully validated BibTeX; this workflow produces final.bib ready for submission.
