fabricatio-typst 0.1.30

A Python library for generating, validating and converting academic content using Typst format.

fabricatio-typst

MIT

Academic article generation and Typst document tooling, built on Fabricatio's event-based agent framework.

Installation

pip install fabricatio[typst]

For RAG-backed article writing:

pip install fabricatio[typst,rag]

Overview

Two layers compose this package:

Rust extension (fabricatio_typst.rust) — performance-critical utilities:

• TeX-to-Typst math conversion (raw LaTeX and delimited $/$$/\(/\[ math)
• BibTeX bibliography management with fuzzy citation lookup
• Typst comment manipulation and YAML front-matter handling
• Markdown section extraction

Python layer — agent-based academic content generation:

• Extract paper essences and generate structured research proposals
• Build hierarchical article outlines and generate full content
• RAG-backed article writing with citation-aware iterative retrieval
• Store and query article chunks in LanceDB

Rust API

from fabricatio_typst.rust import (
    BibManager,
    tex_to_typst,
    convert_all_tex_math,
    comment,
    uncomment,
    strip_comment,
    split_out_metadata,
    to_metadata,
    extract_body,
    replace_thesis_body,
    extract_sections,
    fix_misplaced_labels,
)

TeX Conversion

Function	Description
tex_to_typst(string)	Convert raw LaTeX code to Typst math
convert_all_tex_math(string)	Convert $, $$, \(, \[ math delimiters to Typst $

tex_to_typst(r"\frac{1}{\sqrt{x^2 + 1}}")
# => '#frac(1, sqrt(x^2 + 1))'

convert_all_tex_math("Einstein's $E = mc^2$ is famous.")
# => "Einstein's $E = m c^2$ is famous."

Comment Utilities

Function	Description
comment(string)	Prepend // to every line
uncomment(string)	Remove // prefix from every line
strip_comment(string)	Strip leading and trailing comment lines

BibManager

Loads a BibTeX bibliography file and provides fuzzy citation lookup.

bib = BibManager("references.bib")
cite_key = bib.get_cite_key_by_title("Attention Is All You Need")
# => "vaswani2017attention"

key = bib.get_cite_key_fuzzy("transformers attention mechanism")
authors = bib.get_author_by_key(key)
abstract = bib.get_abstract_by_key(key)

Method	Description
get_cite_key_by_title(title)	Exact-match citation key by title
get_cite_key_by_title_fuzzy(title)	Fuzzy-match citation key by title
get_cite_key_fuzzy(query)	Fuzzy-match citation key by arbitrary query
list_titles(is_verbatim=False)	List all bibliography titles
get_author_by_key(key)	Authors for a citation key
get_year_by_key(key)	Publication year for a citation key
get_abstract_by_key(key)	Abstract for a citation key
get_title_by_key(key)	Title for a citation key
get_field_by_key(key, field)	Arbitrary BibTeX field for a citation key

Document Utilities

Function	Description
split_out_metadata(string)	Extract YAML front matter as (metadata, rest)
to_metadata(data)	Serialize Python object to YAML comment block
extract_body(string, wrapper)	Extract content between wrapper markers
replace_thesis_body(string, wrapper, new_body)	Replace content between wrapper markers
extract_sections(string, level=1, section_char="#")	Parse markdown sections at given header level
fix_misplaced_labels(string)	Move \<label\> tags outside display math blocks

Python Models

Hierarchical article representation from proposal through completed paper:

Model	Description
ArticleProposal	Research proposal: problems, approaches, methods, aims, keywords
ArticleEssence	Semantic fingerprint: title, authors, equations, figures, contributions
ArticleOutline	Hierarchical outline: chapters → sections → subsections
ArticleChapterOutline	Chapter-level outline node
ArticleSectionOutline	Section-level outline node
ArticleSubsectionOutline	Subsection-level outline node
Article	Complete paper with full content and validation
ArticleChapter	Chapter with sections
ArticleSection	Section with subsections
ArticleSubsection	Subsection with paragraphs
Paragraph	Structured paragraph blueprint with word count
ArticleChunk	LanceDB-storable chunk with bibtex metadata
ArticleEssenceStorable	ArticleEssence with LanceDB storage capability
CitationManager	Deduplicated, iteratively expanded citation set for RAG
ChunkKwargs	TypedDict for chunking parameters

Actions

Each action is a Fabricatio Action — an async callable unit in the agent workflow. Available actions:

Content Extraction

Action	Description
ExtractArticleEssence	Extract article essences from files on disk
FixArticleEssence	Fix extracted essences using BibTeX reference data
ExtractOutlineFromRaw	Parse an outline from raw text

Generation

Action	Description
GenerateArticleProposal	Generate a research proposal from a briefing
GenerateInitialOutline	Build an outline from a proposal
GenerateArticle	Generate full article content from an outline
LoadArticle	Load a complete article from outline + Typst code
WriteChapterSummary	Write summaries for each chapter
WriteResearchContentSummary	Write a research content summary

RAG-Backed Writing

Action	Description
WriteArticleContentRAG	Write article content with citation-aware RAG
ArticleConsultRAG	Retrieve relevant citations for article sections
TweakArticleLancedbRAG	Refine article content using LanceDB RAG
ChunkArticle	Split an article into storeable chunks
StoreArticleEssence	Store article essences into LanceDB

Citation Capability

CitationLancedbRAG — citation-aware iterative search that expands queries over multiple rounds and deduplicates by bibtex key.

Workflows

Pre-composed action pipelines:

Workflow	Steps
WriteOutlineCorrectedWorkFlow	GenerateArticleProposal → GenerateInitialOutline → dump output
StoreArticle	ExtractArticleEssence → StoreArticleEssence

Usage:

from fabricatio_typst.actions.article import GenerateArticleProposal, GenerateInitialOutline
from fabricatio_typst.models.article_proposal import ArticleProposal
from fabricatio_typst.models.article_outline import ArticleOutline

async def generate_outline():
    proposal: ArticleProposal = await GenerateArticleProposal()._execute(
        article_briefing="Quantum error correction in near-term devices"
    )
    print(f"Proposal: {proposal.title}")

    outline: ArticleOutline = await GenerateInitialOutline()._execute(
        article_proposal=proposal
    )
    print(f"Outline chapters: {len(outline.chapters)}")

Dependencies

• fabricatio-core — agent framework and core models
• fabricatio-tool — filesystem utilities
• fabricatio-capabilities — capability mixins (Extract, Censor, etc.)

Optional for article generation workflows:

• fabricatio-actions — output dumping actions
• fabricatio-improve, fabricatio-rule — content improvement and validation

Optional for RAG features:

• fabricatio-lancedb — vector storage and retrieval

CLI

The package includes the ttm (TeX to Typst Math) binary:

# Convert math in a file
ttm file input.md -o output.md

# Convert raw LaTeX string
ttm raw "\frac{a}{b}"

# Convert a string with math delimiters
ttm string "The formula $\int_0^\infty e^{-x} dx$ is useful."

Configuration

TypstConfig exposes tunable settings via the Fabricatio config system:

from fabricatio_typst.config import typst_config

typst_config.paragraph_sep         # "// - - -"
typst_config.article_wrapper       # "// =-=-=-=-=-=-=-=-=-="
typst_config.chap_summary_template # "built-in/chap_summary"

License

MIT — see LICENSE