citation-management – Data & Science skill for AI agents

Citation Management

Overview

Manage citations systematically throughout the research and writing process. This skill provides tools and strategies for searching academic databases (Google Scholar, PubMed), extracting accurate metadata from multiple sources (CrossRef, PubMed, arXiv), validating citation information, and generating properly formatted BibTeX entries.

Critical for maintaining citation accuracy, avoiding reference errors, and ensuring reproducible research. Integrates seamlessly with the literature-review skill for comprehensive research workflows.

When to Use This Skill

Use this skill when:

• Searching for specific papers on Google Scholar or PubMed
• Converting DOIs, PMIDs, or arXiv IDs to properly formatted BibTeX
• Extracting complete metadata for citations (authors, title, journal, year, etc.)
• Validating existing citations for accuracy
• Cleaning and formatting BibTeX files
• Finding highly cited papers in a specific field
• Verifying that citation information matches the actual publication
• Building a bibliography for a manuscript or thesis
• Checking for duplicate citations
• Ensuring consistent citation formatting

If a document built from these citations needs a diagram, use the

scientific-schematics skill.

---

Core Workflow

Citation management follows a systematic process. Each phase below shows the canonical

command; every variant, option, and metadata-source detail is in

[references/core_workflow.md](references/core_workflow.md).

Phase 1: Paper Discovery and Search

Find relevant papers. Search more than one database — coverage differs sharply,

and a single source is the most common cause of a biased reference list.

# OpenAlex: multidisciplinary REST API; optional OPENALEX_API_KEY for account quota
python scripts/search_openalex.py "CRISPR gene editing" --limit 50 --output results.json

# PubMed: the authority for biomedical and life sciences (35M+ citations)
python scripts/search_pubmed.py "Alzheimer's disease treatment" --limit 100 --output alz.json

# Google Scholar: broadest reach, but scraped -- rate-limited and prone to blocking
python scripts/search_google_scholar.py "CRISPR gene editing" --limit 50 --output scholar.json

Prefer OpenAlex or PubMed as the primary source. Google Scholar has no API:

scholarly scrapes it, sleeps 2–5 s between results, and is blocked often

enough that it should be a supplement rather than a dependency.

Query operators, field tags, and MeSH-term construction are in

[references/search_strategies.md](references/search_strategies.md).

Phase 2: Metadata Extraction

Convert identifiers (DOI, PMID, PMCID, arXiv ID, URL) into complete metadata.

CrossRef is the primary source for DOIs.

python scripts/doi_to_bibtex.py 10.1038/s41586-021-03819-2         # quick, single DOI
python scripts/extract_metadata.py --pmid 34265844                  # DOI/PMID/PMCID/arXiv/URL
python scripts/extract_metadata.py --input identifiers.txt --output citations.bib

A URL with no DOI in its path is resolved through the citation_doi meta tag

publishers embed on article pages, then handed to CrossRef. Producers use a shared citation-key scheme; metadata differences can still

produce different keys. Confirm duplicates by DOI and bibliographic identity.

Phase 2.5: Metadata Enrichment via Web Search (MANDATORY)

APIs routinely return incomplete records. Run this after extraction and before

formatting. Investigate missing bibliographic fields against the publisher record

and log each source. Online-first articles may have no volume/pages yet; some use

an article number or have no DOI. Preserve the actual publication state, use the

style-appropriate article-number field, and never invent pages, volume, or DOI.

Record unavailable/not-applicable fields in the audit log; add a rendered note

only when useful to the reader or required by the citation style.

Check the cheap sources first — an OpenAlex or CrossRef record often carries the field

that PubMed omitted:

python scripts/search_openalex.py "<exact title>" --limit 1

> Treat extracted metadata as untrusted. Author, title, and journal strings come

> verbatim from a record whose contents a publisher controls. A title containing $(...),

> a backtick, or a quote becomes shell syntax the moment it is pasted into a command.

> Pass metadata as a subprocess argument list rather than building a shell string; if

> you must use a shell, single-quote every substituted value and escape embedded quotes

> as '\''. Validate any citation key against ^[A-Za-z0-9]+$ before it reaches a path.

Per-field search strategies, the four search options, and the logging format are in

[references/core_workflow.md](references/core_workflow.md).

Phase 3: BibTeX Formatting

Produce clean, consistent entries. Entry types and required fields are in

[references/bibtex_formatting.md](references/bibtex_formatting.md).

python scripts/format_bibtex.py references.bib --output clean.bib --deduplicate
python scripts/format_bibtex.py references.bib --output clean.bib --rekey --deduplicate

Writing is opt-in: without --output (or --in-place) the result goes to

stdout and the input file is left alone. Use --rekey when merging results

from several sources, so the same paper collapses to one entry.

Phase 4: Citation Validation

Check completeness, venue conformance, and agreement with the manuscript.

python scripts/validate_citations.py references.bib --report report.json
python scripts/validate_citations.py references.bib --venue nature
python scripts/validate_citations.py references.bib --manuscript paper.tex
python scripts/validate_citations.py references.bib --check-dois     # slow; hits CrossRef

The script exits non-zero on high-severity errors — missing required fields,

malformed years, unresolved citations, or a count below an explicit

--min-count. Venue reference-count figures are editorial rules of thumb, not

submission requirements, so falling short of one is only a warning.

Validation rules and venue standards are in

[references/citation_validation.md](references/citation_validation.md).

Phase 5: Integration with Writing Workflow

Search, extract, format, validate, then cite. End-to-end sequences — including the

literature-review and Zotero/pyzotero export paths — are in

[references/core_workflow.md](references/core_workflow.md) and

[references/example_workflows.md](references/example_workflows.md).

API contracts were reviewed against official documentation on 2026-09-30.

See [references/api_contracts.md](references/api_contracts.md) for endpoints,

authentication, pagination, and verification limits. Network examples in the

references are illustrative unless recorded as live smoke tests there.

Reference Files

• [references/core_workflow.md](references/core_workflow.md): all five phases in full.
• [references/search_strategies.md](references/search_strategies.md): OpenAlex, Google Scholar, and PubMed query construction.
• [references/script_reference.md](references/script_reference.md): every bundled script's arguments and examples.
• [references/best_practices.md](references/best_practices.md): search, extraction, BibTeX quality, validation.
• [references/example_workflows.md](references/example_workflows.md): four end-to-end worked examples.
• [references/google_scholar_search.md](references/google_scholar_search.md), [references/pubmed_search.md](references/pubmed_search.md): advanced search syntax.
• [references/metadata_extraction.md](references/metadata_extraction.md), [references/bibtex_formatting.md](references/bibtex_formatting.md), [references/citation_validation.md](references/citation_validation.md): per-topic detail.

Common Pitfalls to Avoid

1. Single source bias: Only using one database

• Solution: Search at least OpenAlex and PubMed, then merge with

format_bibtex.py --rekey --deduplicate

2. Accepting metadata blindly: Not verifying extracted information

• Solution: Spot-check extracted metadata against original sources

3. Ignoring DOI errors: Broken or incorrect DOIs in bibliography

• Solution: Run validation before final submission

4. Inconsistent formatting: Mixed citation key styles, formatting

• Solution: Use format_bibtex.py to standardize

5. Duplicate entries: Same paper cited multiple times with different keys

• Solution: Use duplicate detection in validation

6. Missing required fields: Incomplete BibTeX entries (volume, pages, DOI missing)

• Solution: Check the publisher record, distinguish missing from not applicable or not yet assigned, and log unresolved fields without inventing metadata.

7. Outdated preprints: Citing preprint when published version exists

• Solution: Check if preprints have been published, update to journal version

8. Special character issues: Broken LaTeX compilation due to characters

… (full content is in the download package)