wyh0626/resume-optimizer

What it solves

Most resumes list duties ("I was responsible for X") rather than results. This project is a skill that finds the most damaging problems in a resume, rewrites vague or duty-focused wording into achievement-oriented language, surfaces missing metrics and business value, and adjusts emphasis to match a target job description. It also flags common risks such as outsourcing experience, toy projects, missing quantifiable results, and inflated technical wording — and can generate a compressed one-page version on request.

How it works

The project is packaged as a "skill" that plugs into AI coding agents like Claude Code, Codex, Cursor, GitHub Copilot, Windsurf, Gemini, Cline, OpenCode, and Zed (20+ supported clients, installed via npx skills add wyh0626/resume-optimizer or a manual git clone). Once installed, you can invoke it explicitly ("Use $resume-optimizer to audit my resume...") or trigger it naturally with a request like "帮我审计这份简历，并按高级后端工程师 JD 优化重点". The skill works through a defined workflow: it produces a 30-second summary, a list of critical issues, value extraction and quantification suggestions, a revision strategy, rewritten resume snippets or a full draft, a check for whether a one-page version is needed, and a next-step action list. It uses reference files (audit checklist, narrative tools, one-page resume guidance, red flags) and an OpenAI agent config to drive the optimization.

For final formatting, the README recommends pairing this content-focused skill with the Kami typesetting system. Resume Optimizer handles what to write and how to phrase it; Kami handles layout, fonts, spacing, and PDF export. The README includes detailed guidance on switching Kami's default warm paper background to pure white for safer printing, with CSS variable adjustments.

Who it's for

Job seekers preparing technical resumes — especially those applying for experienced (社招) roles, students writing project descriptions for campus recruiting, people who need to tailor a resume to a specific JD before applying, anyone compressing a two-page-plus resume to one page, or those doing a pre-interview self-check. It also helps people fix narratives around outsourcing experience, career transitions, or employment gaps.

Highlights

• Focused single-purpose design: audits and rewrites resumes for persuasiveness, not generic advice
• Converts duty descriptions ("did X") into achievement descriptions ("accomplished X") without inventing data
• Proactively uncovers quantitative results, business value, and actual deliverables
• Tailors keywords, project emphasis, and career narrative to a target job description
• Flags common red flags: outsourcing experience, toy projects, missing metrics, inflated tech labels
• Can produce a compressed one-page resume version on demand
• Easy install via npx skills add for 20+ AI coding agents, or manual clone
• Strongly recommends pairing with Kami (a document typesetting system) for clean PDF output, with print-safe white background advice