Citation Management
Citation Management is a skill in K-Dense's Claude Scientific Writer (MIT) that handles the unglamorous part of research writing: finding papers, pulling accurate metadata, producing clean BibTeX, and checking the bibliography before you submit. Version 1.6 ships six Python scripts for that pipeline.
Key Features
- Search:
search_google_scholar.pyandsearch_pubmed.pyquery Google Scholar and PubMed (via NCBI E-utilities) and save results as JSON. - Metadata extraction:
doi_to_bibtex.pyconverts a single DOI, andextract_metadata.pyhandles DOIs, PMIDs, arXiv IDs, or URLs, singly or from a list. CrossRef is the primary source for DOIs. - Mandatory enrichment: any
@articlemissing volume, pages, or DOI after extraction must be filled by web search (through the Parallel web skill) or given anoteexplaining the gap. - Formatting and validation:
format_bibtex.pycleans and deduplicates a.bibfile, andvalidate_citations.pychecks completeness, venue conformance (for example--venue nature), and agreement with a LaTeX manuscript. - Safety note: the skill tells agents to treat publisher metadata as untrusted input, pass it as argument lists rather than shell strings, and validate citation keys against
^[A-Za-z0-9]+$.
Use Cases
- Converting a reference list of DOIs and PMIDs into one consistent BibTeX file.
- Checking a thesis bibliography for duplicates, broken DOIs, and missing fields.
- Replacing preprints with their published journal versions before submission.
Pricing and Access
Free under MIT. CrossRef, doi.org, and arXiv are queried without credentials. Optional NCBI_EMAIL and NCBI_API_KEY identify you to NCBI and raise rate limits. OPENROUTER_API_KEY is only for optional diagrams.
Getting Started
- Install the plugin in Claude Code:
/plugin marketplace add https://github.com/K-Dense-AI/claude-scientific-writer, then/plugin install claude-scientific-writer. pip install requests bibtexparser biopython(addscholarlyfor Google Scholar).- Try
python scripts/doi_to_bibtex.py 10.1038/s41586-021-03819-2, thenpython scripts/validate_citations.py references.bib --report report.txt.
Limitation: Google Scholar has no official API, so its search relies on scraping libraries and can be rate-limited or blocked. Extracted metadata can still be wrong; the skill asks you to spot-check against the source.
Frequently Asked Questions
Does it replace Zotero?
No. It produces and cleans BibTeX; the skill's references describe exporting to Zotero via pyzotero rather than replacing a reference manager.
Can it check citations inside my paper?
Yes, validate_citations.py --manuscript paper.tex compares the .bib file with citations used in the LaTeX source.
Alternatives
- Literature Review Skill: the systematic search and synthesis this skill is designed to support.
- Research Lookup Skill: builds a verified reference packet, including a
references.bib. - Scholar Evaluation Skill: judges the quality of the work you cite.
Conclusion
Citation Management is a practical fit if you write in LaTeX and want scripted BibTeX extraction and validation inside Claude Code. It will not judge whether a source supports your claim; that check stays with you. More research skills are in the skills hub.
Comments
No comments yet. Be the first to comment!
Related Tools
Related Insights

Anthropic Subagent: The Multi-Agent Architecture Revolution
Deep dive into Anthropic multi-agent architecture design. Learn how Subagents break through context window limitations, achieve 90% performance improvements, and real-world applications in Claude Code.
Skills + Hooks + Plugins: How Anthropic Redefined AI Coding Tool Extensibility
An in-depth analysis of Claude Code's trinity architecture of Skills, Hooks, and Plugins. Explore why this design is more advanced than GitHub Copilot and Cursor, and how it redefines AI coding tool extensibility through open standards.
Claude Code's Next Frontier: Not Code, But Your Local Obsidian Knowledge Base
Explore how Obsidian + Claude Code transforms from a knowledge management tool into your private AI assistant. Complete guide including obsidian-skills, Claudian plugin, Claudesidian template, and best practices for achieving both data privacy and AI capabilities.