← 返回 Skills 市场

Hr Resume Evaluation

Name: Hr Resume Evaluation
Author: debtvc2022

作者 CUPVC · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

总下载

当前安装

版本数

在 OpenClaw 中安装

/install hr-resume-evaluation

功能描述

Use when a recruiter wants to batch-evaluate resumes, run general resume quality checks, optionally match resumes to a JD and company background, and produce...

使用说明 (SKILL.md)

HR Resume Evaluation

Core Rule

Use this Skill only as a recruiter review aid. Never present scores as final hiring, rejection, elimination, or admission decisions.

Before Use

Put resumes in a local input folder. Supported resume formats: Markdown, TXT, PDF, DOCX.
Optionally prepare a JD file and company background file in Markdown or TXT.
Set the API key environment variable configured in config/model.yaml, default DEEPSEEK_API_KEY.
Read references/scoring-rubric.md, references/role-family-taxonomy.md, references/compliance-policy.md, and references/output-schema.md.

Command

python scripts/evaluate_resumes.py `
  --resumes input/resumes `
  --jd input/jd.md `
  --company input/company.md `
  --output reports `
  --config config/evaluation.yaml `
  --model-config config/model.yaml

Omit --jd and --company for general resume quality review only.

Required Outputs

One Markdown report per successfully evaluated resume.
One batch summary Markdown report.
One CSV summary file.
Clear failure records for unreadable or invalid files.

Safety Rules

Do not score protected or job-irrelevant traits.
Do not infer missing resume facts.
Do not perform background investigation.
Do not automatically reject or approve candidates.
Treat model output as advisory until reviewed by a human recruiter.

安全使用建议

Install only if your organization permits sending resumes and recruiting materials to the configured model provider. Review config/model.yaml, confirm the endpoint and API key policy, disable raw JSON output if unnecessary, and avoid using the reports as automatic hiring or rejection decisions.

能力标签

requires-sensitive-credentials

能力评估

⚠ Purpose & Capability

The stated purpose is recruiter-side resume review, and the scripts match that purpose by parsing resumes, asking a model to score them, and writing reports; the concern is that full resume, JD, and company text can be sent to the configured DeepSeek-compatible API, which is sensitive HR data handling.

⚠ Instruction Scope

The command is user-directed and scoped to provided input/output paths, but SKILL.md does not plainly tell users that candidate materials are transmitted to a third-party model endpoint or require approval before doing so.

✓ Install Mechanism

No hidden installer, startup hook, package mutation, or automatic execution mechanism was found; dependencies are ordinary parsing, HTTP, YAML, templating, and test packages.

⚠ Credentials

Reading local resumes, using an API key, network access, and writing reports are proportionate to the feature, but the privacy controls are thin for likely PII and confidential recruiting documents.

ℹ Persistence & Privilege

The skill writes Markdown, CSV, and by default raw JSON evaluation outputs to the chosen reports directory; there is no evidence of background persistence, privilege escalation, or unrelated credential access.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install hr-resume-evaluation
安装完成后，直接呼叫该 Skill 的名称或使用 /hr-resume-evaluation 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial release of HR Resume Evaluation skill. - Enables batch evaluation of resumes for recruiters, supporting Markdown, TXT, PDF, and DOCX formats. - Optionally matches resumes to job description (JD) and company background to enhance review context. - Produces evidence-backed evaluation reports in Markdown and summary results in CSV, with clear handling of failures. - Strictly advisory: scores must not be used as final hiring or rejection decisions. - Includes comprehensive safety and compliance guidelines to ensure ethical use.

元数据

Slug hr-resume-evaluation

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题