{"skills":[{"name":"Leadership","esco_label":"leadership","type":"soft","confidence":0.95},{"name":"Mentoring","esco_label":"mentoring","type":"soft","confidence":0.95}]}
curl --location --request POST 'https://zylalabs.com/api/13176/multilingual+skills+extraction+api/26723/parse+skills?text=led the team and mentored 3 juniors' --header 'Authorization: Bearer YOUR_API_KEY'
注册后,每个开发者都会被分配一个个人 API 访问密钥,这是一个唯一的字母和数字组合,用于访问我们的 API 端点。要使用 Multilingual Skills Extraction API 进行身份验证,只需在 Authorization 标头中包含您的 bearer token。
| 标头 | 描述 |
|---|---|
授权
|
必需
应为 Bearer access_key. 订阅后,请查看上方的"您的 API 访问密钥"。
|
无长期承诺。随时升级、降级或取消。 免费试用包括最多 50 个请求。
(年度计费可节省 2 个月 🎉)
领先企业的信赖之选
Skills Extraction API turns any job posting or resume into a clean, structured list of professional skills, in any language.
Unlike keyword-based parsers, it uses an LLM to understand context, so it catches IMPLICIT soft skills ("led the team"
becomes Leadership, "mentored juniors" becomes Mentoring) and works on non-English text (Korean, Spanish, Japanese, etc.)
where simpler parsers return an empty result.
For each skill you get:
- name: canonical, searchable name (for example "js" becomes "JavaScript", "k8s" becomes "Kubernetes")
- esco_label: standard ESCO/O*NET-style label
- type: "hard" or "soft"
- confidence: a 0 to 1 score
One endpoint: POST /parse-skills with { "text": "..." } returns { "skills": [...] }
Built for ATS enrichment, job-candidate matching, resume parsing, talent analytics, and recruiting automation. No
preprocessing and no model setup: paste raw text, get normalized JSON back.
The API extracts both hard and soft skills from text, including implicit soft skills such as leadership and mentoring, by understanding the context in which they are mentioned.
The API returns a clean, structured JSON format that includes fields such as 'name' (canonical name), 'esco_label' (standard label), 'type' (hard or soft), and 'confidence' (a score from 0 to 1).
The API can be used for ATS enrichment, job-candidate matching, resume parsing, talent analytics, and recruiting automation, making it valuable for HR and recruitment professionals.
The core value proposition lies in its ability to provide accurate, normalized skill extraction from diverse text inputs, enhancing recruitment processes and improving talent matching through structured data.
Any language the underlying LLM understands, which covers most major world languages. Skill names are always normalized to canonical English.
The API returns a structured JSON object containing a list of normalized professional skills extracted from the input text. Each skill entry includes its canonical name, ESCO label, type (hard or soft), and a confidence score.
The key fields in the response data include 'name' (the normalized skill name), 'esco_label' (the standard label), 'type' (indicating if the skill is hard or soft), and 'confidence' (a score from 0 to 1 representing the extraction certainty).
The response data is organized as a JSON object with a 'skills' array. Each element in the array represents a skill, containing fields for name, ESCO label, type, and confidence score, allowing for easy parsing and utilization.
The endpoint provides information on professional skills extracted from job postings or resumes, including both hard skills (like programming languages) and soft skills (like teamwork), along with their respective confidence scores.
用户可以通过在POST请求中向/parse-skills端点提供原始文本输入来自定义他们的数据请求。API会自动处理文本,而不需要额外的参数或预处理
典型的用例包括增强申请人跟踪系统(ATS)改善求职者匹配自动化简历解析进行人才分析和简化人力资源专业人员的招聘流程
Data accuracy is maintained through the use of a large language model (LLM) that understands context, allowing it to extract implicit skills effectively. Continuous training and updates to the model help ensure high-quality skill extraction.
If users receive partial or empty results, they should check the input text for clarity and context. Providing more detailed or structured text can improve extraction results, as the API relies on context to identify skills accurately.