kapi exec qa

$ kapi qa [files...]

Check translation quality with deterministic rules or an LLM (select with --mode)

Commandexec qa

Try it (demo)

Demo mode — illustrative output from a built-in stub, not a real model. Install the CLI to run with your own API key.

$ kapi qa app.xliff --target-lang fr

Flags

Flag	Type	Default	Description
--absolute-max-char-length	int	255	Maximum number of characters allowed in any target segment
--api-key	string		API key for the AI provider
--charset	string	ISO-8859-1	Name of the character set encoding to check against (e.g. ISO-8859-1)
--check-absolute-max-char-length	bool	false	Flag target segments that exceed an absolute character count limit
--check-charset	bool	false	Warn if a target character is not included in the specified character set encoding
--check-code-difference	bool	true	Verify that target segments have the same inline codes as source segments
--check-corrupted-characters	bool	true	Check for patterns indicating encoding corruption (mojibake
--check-double-spaces	bool	true	Check for double spaces in target text
--check-doubled-word	bool	true	Check for consecutive repeated words in target text
--check-empty-source	bool	true	Check for non-empty target when source is empty
--check-empty-target	bool	true	Check for empty target when source has content
--check-leading-whitespace	bool	true	Check for leading whitespace mismatches between source and target
--check-max-char-length	bool	true	Flag targets longer than a percentage of source character length
--check-max-words	bool	false	Flag target segments that exceed an absolute word count limit
--check-min-char-length	bool	true	Flag targets shorter than a percentage of source character length
--check-patterns	bool	true	Verify that source patterns have expected corresponding content in the target
--check-source-inconsistency	bool	false	Flag when different source texts share the same translation
--check-span-constraints	bool	true	Check non-deletable and non-cloneable span constraint violations
--check-target-inconsistency	bool	false	Flag when the same source text is translated differently across the run
--check-target-same-as-source	bool	true	Check when target text is identical to source text
--check-terminology	bool	false	Enable terminology checks
--check-trailing-whitespace	bool	true	Check for trailing whitespace mismatches between source and target
--checks	stringSlice		Quality checks to perform (e.g. terminology fluency accuracy consistency)
--concurrency, -j	int	0	max parallel files (0 = auto)
--consistency-case-sensitive	bool	true	Whether consistency comparison is case-sensitive
--credential	string		saved credential name to use (see 'kapi credentials list')
--doubled-word-exceptions	string	sie;vous;nous	Semicolon-separated list of words allowed to repeat (e.g. sie;vous;nous)
--encoding, -e	string	UTF-8	input file encoding
--fail-on-unknown	bool	false	exit with error if any file cannot be processed (default: skip with warning)
--forbidden-chars	string		Characters that must not appear in target text (e.g. {}[])
--format, -f	string		override input format detection
--json	bool	false	output results as JSON
--map, -m	stringArray		map glob pattern to format (e.g. '*.docx=okf_openxml:test')
--max-char-length-above	int	200	Maximum allowed percentage of source length for long text
--max-char-length-below	int	350	Maximum allowed percentage of source length for short text
--max-char-length-break	int	20	Character count above which text is considered long for the maximum length check
--max-words	int	0	Maximum number of words allowed in any target segment
--min-char-length-above	int	45	Minimum required percentage of source length for long text
--min-char-length-below	int	30	Minimum required percentage of source length for short text
--min-char-length-break	int	20	Character count above which text is considered long for the minimum length check
--mode	string	rules	How to check quality: deterministic local rules, or an AI provider's review (Deterministic rules, AI review)
--model	string		AI model name
--no-warn	bool	false	suppress warnings for skipped files
--output, -o	string		output path template (variables: {dir}, {name}, {ext}, {lang})
--output-dir	string		write outputs under DIR/{lang}/ (default: beside the input, mirroring its locale layout)
--pack	bool	false	when transforming a .kpz, also eject the result to the .kpz (auto-pack)
--parallel-blocks	int	0	fan out block processing across N goroutines (0 = off)
--progress, -p	bool	false	show progress bar
--provider	string	anthropic	AI provider (Anthropic, OpenAI, Gemini, Azure OpenAI, Claude Code, Ollama, Demo (illustrative))
--required-chars	string		Characters that must appear in target if present in source (e.g. punctuation)
--source-lang	string	en	source language (e.g. en, en-US)
--strict	bool	false	alias for --fail-on-unknown
--strict-code-order	bool	false	Flag differences when codes appear in a different order between source and target
--target-lang	string		target language (e.g. fr, de-DE)
--target-same-as-source-with-codes	bool	true	Include inline codes when comparing source and target for identity
--target-same-as-source-with-numbers	bool	true	Include number-only segments in same-as-source comparison
--trace	string		write flow trace JSON to file (for flow visualization)

More examples

$ kapi qa app.xliff --target-lang fr --provider anthropic

$ kapi qa app.xliff --target-lang de --json

← Back to the Command Reference