菜单
Extract a PDF, DOCX, pasted resume, or RenderCV YAML into a structured RenderCV-compatible candidate profile with source-backed facts and confidence. Use when importing, normalizing, or validating a resume before job matching or tailoring.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
Propose reviewable structured edits to a local master resume from a user conversation, optionally using one job or a server-resolved local market sample as wording and prioritization context.
Classify how a trusted master resume covers a fixed list of aggregated market skills using verifiable resume paths and confirmed fact IDs.
Classify how a trusted resume covers a fixed set of job requirements using explicit resume paths and confirmed fact IDs.
Generate a focused interview preparation plan from aggregated local job-market statistics and, when available, confirmed resume facts. Use when a candidate requests interview skill priorities, personal capability gaps, preparation actions, project stories, or practice questions for the jobs stored in this app.
Extract clean, source-grounded job, company, and business-registration fields from a noisy BOSS job-detail page captured as plain text.
Evaluate a job posting against a confirmed resume and stated job preferences, producing evidence-backed fit dimensions, hard constraints, strengths, gaps, confidence, and a recommendation. Use when ranking jobs or opening a job's match analysis.
| name | resume-extraction |
| description | Extract a PDF, DOCX, pasted resume, or RenderCV YAML into a structured RenderCV-compatible candidate profile with source-backed facts and confidence. Use when importing, normalizing, or validating a resume before job matching or tailoring. |
Convert the supplied resume text into the JSON contract below. Treat the resume as the only source of biographical truth.
confirmed to false.rawText as the source whether it came from a PDF text layer, DOCX, or page-by-page visual transcription. Page markers such as --- Page 2 --- are source identifiers, not resume content.startDate and endDate. For 2024.12 - 至今, return startDate: "2024.12" and endDate: "至今". Never place a full range in one field and never add a leading dash.本科, 硕士, 博士, or 其他. For any other degree, set degree to 其他 and preserve the exact source wording in degreeDetail.confirmed: false; only the user can confirm imported facts.Return one ResumeProfile object:
{
"id": "resume-master",
"name": "",
"headline": "",
"email": "",
"phone": "",
"location": "",
"website": "",
"summary": "",
"templateId": "ai-engineering",
"professionalSkills": [{
"id": "",
"label": "核心方向",
"items": [""]
}],
"experiences": [{
"id": "",
"company": "",
"position": "",
"location": "",
"startDate": "",
"endDate": "",
"highlights": [""]
}],
"education": [{
"id": "",
"institution": "",
"area": "",
"degree": "",
"degreeDetail": "",
"startDate": "",
"endDate": "",
"highlights": []
}],
"projects": [{
"id": "",
"name": "",
"summary": "",
"startDate": "",
"endDate": "",
"highlights": []
}],
"certifications": [{
"id": "",
"name": "",
"issuer": "",
"date": ""
}],
"facts": [{
"id": "fact-stable-id",
"category": "identity|experience|education|skill|project|certification|other",
"value": "",
"source": "section and source excerpt identifier",
"confidence": 0.0,
"confirmed": false
}],
"preferences": {
"targetRoles": [],
"cities": [],
"remotePreference": "flexible",
"energizingTasks": [],
"drainingTasks": [],
"hardConstraints": []
},
"sourceFileName": "",
"updatedAt": "ISO-8601",
"version": 1
}
Use a confidence of at least 0.9 only for directly readable source claims. Keep user preferences empty; they are collected separately.