← 返回 Skills 市场
flychicken123

HiHired Resume

作者 flychicken · GitHub ↗ · v0.1.0 · MIT-0
cross-platform ⚠ suspicious
42
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install hihired-resume
功能描述
Turn HiHired (hihired.org) into a resume copilot workflow. Use when a user wants help building, importing, rewriting, tailoring, or improving a resume, gener...
使用说明 (SKILL.md)

HiHired Resume

Use this skill when the user wants resume help and HiHired is the right execution path.

What this skill does

This skill packages HiHired's existing product flow into a chat-friendly assistant workflow:

  • build a resume from scratch
  • import a resume from file or pasted text
  • rewrite experience bullets
  • generate or improve a professional summary
  • generate skills from resume context
  • generate a matching cover letter from resume + job description
  • guide the user into the HiHired builder for final review/download

HiHired is not just a static template site. It already has working product capabilities on hihired.org for:

  • resume parsing/import
  • AI experience rewriting
  • AI skills generation
  • AI summary generation
  • AI cover letter generation
  • resume/job-description workflow
  • Chrome autofill for job applications

Default behavior

When the user asks for resume help:

  1. Figure out whether they want one of these intents:
    • start from scratch
    • import existing resume
    • tailor resume to a job
    • improve a section
    • generate cover letter
  2. Collect only the missing inputs.
  3. Do as much useful drafting in chat as possible.
  4. When HiHired's UI would be faster or safer, route them to the builder with a concrete next action.

Do not dump a giant questionnaire up front. Ask for the minimum needed to move the draft forward.

Input collection rules

If starting from scratch

Collect in this order:

  1. target role
  2. 1 to 3 most relevant experiences
  3. skills
  4. education
  5. optional projects

If tailoring to a job

Ask for:

  1. current resume text or resume file
  2. target job description or job URL
  3. optional preferred tone (standard, technical, executive, concise)

If generating a cover letter

Ask for:

  1. resume or relevant background
  2. job description
  3. company name if known

Output style

Prefer practical resume output over abstract advice.

Good:

  • rewritten bullets
  • a ready-to-paste summary
  • a skills list
  • a cover letter draft
  • a short checklist for what to do next in HiHired

Less useful:

  • generic career coaching without concrete edits
  • long theory about resume best practices

HiHired routing guidance

Route the user to HiHired when they need one of these:

  • upload/import their existing resume
  • visually edit a template
  • download the final PDF
  • continue a full resume + cover letter workflow
  • later use Chrome autofill for job applications

Recommended destination:

  • builder: https://hihired.org/builder
  • templates: https://hihired.org/templates
  • cover-letter intent page: https://hihired.org/guides/ai-resume-builder-with-cover-letter

When handing off, say exactly what they should do next, for example:

  • "Open hihired.org/builder, import your PDF, then paste this job description into the Job Description step."
  • "Go to the builder and paste these rewritten bullets into Experience, then use the summary I drafted below."

Conversation patterns

Resume from scratch

  • Ask for target role first.
  • Then ask for the user's recent experience in plain language.
  • Convert that into resume bullets.
  • Draft a summary and skills section.
  • Suggest moving to HiHired builder for final formatting and PDF export.

Resume improvement

  • Ask them to paste the current section.
  • Rewrite it directly.
  • Offer 2 variants when useful: ATS-safe and punchier.

Resume tailoring

  • Compare resume content against the job description.
  • Emphasize missing keywords, measurable impact, and relevance.
  • Rewrite the summary and top bullets first, because that usually gives the biggest lift.

Cover letter

  • Use the same resume context and target role.
  • Keep it specific and not overly formal.
  • Avoid obvious AI fluff.

Constraints

  • Do not invent experience, metrics, degrees, employers, or certifications.
  • If the user gives weak input, produce a clearly marked draft and tell them what assumptions were made.
  • Keep ATS compatibility in mind: standard section names, clear verbs, measurable outcomes, no gimmicks.

API-backed execution

This skill now has a helper script:

  • scripts/hihired_api.py

Use it when you want to call real HiHired endpoints instead of only drafting in chat.

By default the script targets the direct HiHired backend at http://18.190.155.165 because Cloudflare may still block non-browser API signatures on https://hihired.org with Error 1010. Use https://hihired.org for user-facing links and UI handoff; use the direct backend for agent-side API execution. Override with HIHIRED_BASE_URL only when you have confirmed the target accepts script/API calls.

Supported commands

  • parse-resume
  • summary
  • resume-advice
  • auto-skills
  • categorize-skills
  • cover-letter
  • modify-resume
  • template-preference

Typical usage patterns

Parse a resume file:

python scripts/hihired_api.py parse-resume C:\path	o\resume.pdf

Generate a summary:

python scripts/hihired_api.py summary --experience "..." --education "..." --skills @skills.json --job-description "..."

Generate a cover letter from structured resume data:

python scripts/hihired_api.py cover-letter --resume-data @resume.json --job-description "..." --company-name "..."

For PowerShell and other quote-fragile shells, prefer @file.json inputs over inline JSON arrays.

Get resume advice:

python scripts/hihired_api.py resume-advice --resume-data @resume.json --job-description "..."

Infer template preference from natural language:

python scripts/hihired_api.py template-preference --text "I want something modern and clean for a software engineer role"

Rules for using the API helper

  • Prefer the direct backend base URL over the Cloudflare front door when running from agent tooling.
  • Prefer public endpoints that do not require user login.
  • Do not claim a request succeeded unless the script returns a successful response.
  • If the API call fails, report that immediately and fall back to chat drafting when possible.
  • For large structured payloads, write JSON to a temporary file and pass it with @file.json.

References

Read references/hihired-capabilities.md when you need the concrete mapping from user request to existing HiHired features, API endpoint mapping, and suggested handoff language.

安全使用建议
Use the chat-only drafting features freely, but be careful with the API helper. Before allowing uploads or generated API calls, confirm that you are comfortable sending your resume and job details to HiHired, preferably through the official HTTPS website or a verified HTTPS API endpoint.
功能分析
Type: OpenClaw Skill Name: hihired-resume Version: 0.1.0 The skill is classified as suspicious because it instructs the AI agent to bypass Cloudflare security protections by sending sensitive user data (resumes containing PII) to a hardcoded IP address (http://18.190.155.165) instead of the official domain. Furthermore, the helper script `scripts/hihired_api.py` contains an arbitrary file read vulnerability in the `load_json_arg` function, which allows reading any local file via the `@` prefix (e.g., `@/etc/passwd`) and transmitting its contents to the external IP. While these behaviors are framed as functional requirements for the HiHired service, the combination of WAF bypass and potential for arbitrary file exfiltration poses a significant security risk.
能力评估
Purpose & Capability
Resume drafting, parsing, tailoring, cover-letter generation, and HiHired handoff match the stated purpose, but they involve sensitive personal resume and job-search data.
Instruction Scope
The skill tells the agent to use a helper for real HiHired API calls when desired, including local resume-file upload, but does not clearly require explicit user confirmation before sending that data externally.
Install Mechanism
There is no package install or dependency setup, but the metadata says no required binaries/capabilities while SKILL.md documents a Python helper that performs file reads and network API calls.
Credentials
Defaulting to an HTTP raw IP backend instead of the public HTTPS HiHired domain is disproportionate for handling resumes that may include names, contact details, employment history, and education.
Persistence & Privilege
The artifacts do not show credentials, privileged config access, background persistence, local memory storage, deletion, or autonomous long-running behavior.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install hihired-resume
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /hihired-resume 触发
  4. 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
v0.1.0
Initial API-backed HiHired resume copilot skill
元数据
Slug hihired-resume
版本 0.1.0
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

HiHired Resume 是什么?

Turn HiHired (hihired.org) into a resume copilot workflow. Use when a user wants help building, importing, rewriting, tailoring, or improving a resume, gener... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 42 次。

如何安装 HiHired Resume?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install hihired-resume」即可一键安装,无需额外配置。

HiHired Resume 是免费的吗?

是的,HiHired Resume 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

HiHired Resume 支持哪些平台?

HiHired Resume 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 HiHired Resume?

由 flychicken(@flychicken123)开发并维护,当前版本 v0.1.0。

💬 留言讨论