Bib Convert

bib-convert

Convert bibliography files between 5 supported formats:

• BibTeX - Classic LaTeX bibliography format
• BibLaTeX - Extended BibTeX with modern entry types (dataset, software, online)
• CSL JSON - Citation Style Language JSON (universal format)
• RIS - Reference Manager format (tag-based)
• EndNote XML - EndNote reference manager XML format

Usage

Convert bibliography to another format:

npx tsx ${CLAUDE_PLUGIN_ROOT}/scripts/convert.ts input.bib --to=csl-json --output=output.json

Options

• --from=<format> - Source format (auto-detected if omitted)
• --to=<format> - Target format (required)
• --output=<file> - Output file (prints to stdout if omitted)
• --sort - Sort entries by ID
• --indent=<string> - Indentation for formatted output (default: 2 spaces)

Supported Formats

• bibtex - BibTeX (.bib)
• biblatex - BibLaTeX (.bib)
• csl-json - CSL JSON (.json)
• ris - RIS (.ris)
• endnote - EndNote XML (.xml)

Examples

Convert BibTeX to CSL JSON:

npx tsx ${CLAUDE_PLUGIN_ROOT}/scripts/convert.ts references.bib --to=csl-json --output=references.json

Convert RIS to BibTeX with sorting:

npx tsx ${CLAUDE_PLUGIN_ROOT}/scripts/convert.ts papers.ris --to=bibtex --output=papers.bib --sort

Auto-detect source format:

npx tsx ${CLAUDE_PLUGIN_ROOT}/scripts/convert.ts mystery.txt --to=bibtex

Conversion Architecture

The plugin uses a hub-and-spoke architecture with CSL JSON as the intermediate format:

BibTeX ←→ BibLaTeX ←→ CSL JSON ←→ RIS
                          ↕
                      EndNote XML

This ensures maximum fidelity for conversions between related formats while supporting all 5 formats.

Lossy Conversions

Some conversions may be lossy:

To BibTeX (lossy for modern types):

• dataset → @misc (BibTeX has no dataset type)
• software → @misc
• webpage → @misc
• patent → @misc

Warnings are generated for lossy conversions. Original type information is preserved in metadata for round-trip conversions.

To BibLaTeX (no loss): All CSL JSON types map to BibLaTeX types without loss.

Output

The script outputs:

1. Conversion statistics (total, successful, warnings, failed)
2. Warnings for lossy conversions or parse errors
3. Converted bibliography (to file or stdout)

Notes

• Format auto-detection works for most files
• LaTeX special characters are automatically encoded/decoded (ä ↔ "{a})
• Name parsing handles particles (von, van, de) and suffixes (Jr., III)
• Date parsing supports multiple formats (ISO, natural language)