Fiction Number Checker
Extract numerical facts from manuscripts, including focused English number words from zero to ninety-nine in age, count and explicit duration contexts, for human review.
The problem it solves
Fiction Number Checker Fiction Number Checker extracts numerical facts from TXT, Markdown, DOCX, and EPUB manuscripts into a reviewable inventory. It normalizes common ages, money, distances, durations, percentages, and counts, then groups differing values attached to the same inferred or explicitly requested subject.
Who Fiction Number Checker is for
- People who need a small, purpose-built utility instead of a broad platform.
- Users working with fiction, writing, consistency who want the documented v1.1.0 behavior.
- People who prefer an open-source release with visible limitations, source, and license terms.
Intended result
Extract numerical facts from manuscripts, including focused English number words from zero to ninety-nine in age, count and explicit duration contexts, for human review.
This summary is reconciled from the current catalog and repository documentation.
Features in v1.1.0
Capabilities below come from the current project README and release documentation.
Fiction Number Checker extracts numerical facts from TXT, Markdown, DOCX, and EPUB manuscripts into a reviewable inventory. It normalizes common ages, money, distances, durations, percentages, and counts, then groups differing values attached to the same inferred or explicitly requested subject.
The result is a consistency checklist, not an automated verdict.
Verified examples
Screenshots are shown only when the current README references a local source image. Otherwise, repository example files are linked directly.
Verified release evidence
This project does not currently publish a screenshot or dedicated example folder. The current README, tests, changelog, tagged release, and source repository remain the verified examples of its behavior.
Platforms and implementation
The public release claims only the cataloged platforms and technologies.
Supported platforms
- Windows
- macOS
- Linux
Built with
- Python
Changelog
Release notes from the canonical v1.1.0 source tag.
- Require whole-word duration cues and exclude positive-signed number-word phrases.
- Recognize English number words zero through ninety-nine in focused age, count and explicit duration contexts.
- Normalize simple, spaced and hyphenated forms while retaining original excerpts and offsets.
- Skip unsupported larger/signed/decimal phrases and isolated duration idioms; retain read-only format handling.
- Extract numerical facts from TXT, Markdown, DOCX, and EPUB manuscripts.
- Normalize focused ages, money, distances, durations, percentages, and counts.
- Export human-review groups through terminal, Markdown, and JSON reports.
Quick start
The shortest documented path into the current release.
Before running these source commands, download the v1.1.0 source archive, extract it, and open a terminal in the extracted project folder containing README.md. Follow the tagged README for required runtimes and any platform-specific setup.
Python 3.10 or newer is required.
python -m pip install "fiction-number-checker @ https://github.com/loganpendragonmultiverse/fiction-number-checker/releases/download/v1.1.0/fiction_number_checker-1.1.0-py3-none-any.whl" fiction-number-checker manuscript.docx --subject Mara --subject Dorian --markdown number-review.md --json number-review.json fiction-number-checker manuscript.epub --fail-on-review
Supplying subjects is the most precise workflow. Without them, the tool conservatively uses nearby capitalized names and labels unmatched facts “Unassigned.”
Current limitations
These boundaries are part of the product and prevent the page from implying unverified capability.
Examples: Mara is twenty-one years old, Mara waited for two hours, Two years passed, and Mara carried ninety-nine coins. Durations require a preceding cue (for, after, before, within, in, lasted, waited, spent, took, takes, over) or following passed/elapsed; an isolated one day or one second does not create a fact. Count nouns are people/persons, guards, soldiers, coins, ships, books, doors, rooms, horses, children and items, with singular forms where applicable. Unsupported larger, signed, decimal or and phrases are not partially converted. Number-word distance, money and percentage phrases are outside this bounded feature. Original text, offsets, excerpts and subject filters are preserved. These syntactic rules cannot disambiguate every idiom or a name that looks exactly like a supported quantity; review the excerpt.
Automatic subject guesses exclude standalone number-word tokens. Explicit --subject filters still support characters named Seven or similar ambiguous names.
Duration cues match whole words: suffixes such as in in cousin or again do not count. Signed number-word phrases include +twenty and plus twenty and are excluded.
Privacy and licensing
Review the actual data boundary before using a tool with sensitive inputs.
Privacy and safety
Different values are not automatically inconsistent. Characters age, money is spent, distances are estimated, time passes, measurements are rounded, and multiple facts can legitimately occur in one scene. Every review group requires manuscript context.
Files are processed locally and read-only. No manuscript text is uploaded or sent to an API. Reports contain excerpts and should be treated as manuscript material.
Arabic numerals retain their existing detection. Version 1.1 also recognizes English number words zero through ninety-nine, including space or hyphen compounds, in years old ages, explicit duration contexts and a focused count vocabulary. It does not resolve dates, currencies without a dollar sign, chronology, pronouns, or narrative truth.
License and release
Fiction Number Checker is published under MIT. The current cataloged release is v1.1.0, published 2026-10-15.
Related projects
Related projects are selected deterministically from shared catalog tags, category, and implementation technologies—not popularity or paid placement.
Fiction Number Checker v1.1.0
Use the tagged release for downloads and release notes. Use the repository for source, issues, contribution guidance, security reporting, and complete documentation.
Page source: current Forge catalog plus README and CHANGELOG from the canonical GitHub release tag. Fingerprint: c093cee84c301f61.