Endpoint Reference

Skills Extraction & Normalization

Extract and normalize skills from any block of text — a CV section, LinkedIn bio, job description, GitHub README or freeform input. Returns skills grouped by category with canonical, normalized names.

Normalization is what makes skills data usable.

ReactJS, React.js and React all normalize to the canonical React — so your search, filtering and candidate matching operate on consistent values instead of noisy variants.

Technical

Python, Java, SQL, AWS

Frameworks & Libraries

React, Django, Spring, FastAPI

Tools

Git, Docker, Jira, Figma

Soft Skills

Leadership, Communication, Agile

Domain Knowledge

Finance, Healthcare, E-commerce

Languages

French, Arabic, Spanish

POST/v1/skills

Request Parameters

Name	Type	Description
textRequired	string	The text to extract skills from (CV section, bio, job description, README, etc.). Max 50,000 characters.
language	string	Language code for the canonical skill labels (default: 'en').

🌍 Supported Languages50

afAfrikaans

amAmharic

arArabic

bgBulgarian

bnBengali

csCzech

daDanish

deGerman

elGreek

enEnglish

esSpanish

etEstonian

faPersian

fiFinnish

frFrench

guGujarati

heHebrew

hiHindi

hrCroatian

huHungarian

idIndonesian

itItalian

jaJapanese

knKannada

koKorean

ltLithuanian

lvLatvian

mlMalayalam

mrMarathi

msMalay

nlDutch

noNorwegian

plPolish

ptPortuguese

roRomanian

ruRussian

skSlovak

slSlovenian

srSerbian

svSwedish

swSwahili

taTamil

teTelugu

thThai

tlTagalog

trTurkish

ukUkrainian

urUrdu

viVietnamese

zhChinese

Pro Tip

Run candidate resumes and your job descriptions through /v1/skills, then match on the normalized canonical names — no more missed matches because one side wrote 'k8s' and the other 'Kubernetes'.

curl -X POST https://api.evalcv.com/v1/skills \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text":"Built APIs in Python & FastAPI. Used ReactJS and Docker.","language":"en"}'