script: Khoj
family: Indic
type: abugida
whitespace: unspecified
open_type_tag: none
complex_positioning: unknown
unicode: true
diacritics: true
contextual_forms: true
status: Current
baseline: unspecified
ligatures: required
direction: ltr

Khojki is a Brahmi-derived abugida related to the Sharada script. It is used by the Khoja people - an ethnic group of largely Ismaili Shia Muslims - for recording religious literature in the Sindhi language. Khojki has been used since at least the 16th century, originally for manuscripts, but later in printed form also. The script has also been used to write other South Asian languages; however, in recent years its use has declined markedly.

Khojki is structurally similar to other Brahmic scripts. Each of the forty-three characters represents either an independent vowel, or a consonant containing the inherent vowel a. In addition to these, there are eight dependent vowel diacritics. These can attach to a consonant character to replace the inherent vowel with a different one. Other word-forming characters include the virama, which is written after a consonant to silence the vowel entirely, the nukta, which can be written above either vowels or consonants to represent sounds not native to the Sindhi language, and the shadda which represents geminate (long) consonants.

The script was never standardized. In addition, it was used by non-Sindhi speakers to represent other languages. These users adapted the original phonetic values of some letters to fit the phonology of their own languages. This means that there is inconsistency in the representation of some sounds. For example, there are two letters attested in different manuscripts as representing the dental aspirated sound d̪h.

Khojki punctuation is largely borrowed from Latin. However, the danda and double danda are used to mark the end of a sentence, and a word separator is optionally used in between words.