Britfix

Britfix

Convert American spellings to British English. Originally created to fix the US spelling output from LLMs, but useful for anyone who needs British spellings in their documents and code.

Features

- Multiple File Types: Text, markdown, LaTeX, HTML, JSON, and code files
- Smart Code Handling: Only converts comments and docstrings in code files, never string literals or identifiers
- Context-Aware: Preserves quoted text like 'colorScheme' in comments (API references)
- Interactive Mode: Review and approve changes one by one
- Hook Integration: Automatically fix spellings when writing files

Installation

Requires uv and optionally just.

Usage

Options

- --input: Input file(s) or pattern(s). If omitted, reads from stdin
- --interactive, -i: Interactive approval mode (strategy-aware — only suggests changes in regions the strategy considers safe, e.g. comments in code files, prose in markdown). JSON files fall back to non-interactive processing
- --dry-run: Preview changes without modifying files
- --no-backup: Skip backup creation
- --recursive: Process files recursively
- --quiet: Only output corrected text (for pipelines)
- --no-project-ignore: Skip the .britfixignore files found beside the input files. Your user-level ignore file still applies. The hook never passes this

File Type Strategies

Different file types are handled by different strategies, configured in config.json:

An extension not listed above has no strategy, and both the CLI and the hook
skip it rather than guessing. Plain text is not used as a fallback for files on
disk: a .pyi, .yaml, .lua or extensionless file converted as plain text
would have its identifiers and keys rewritten, not just its prose. The CLI
reports each one as a britfix: skipped diagnostic and counts it separately.…