maccydee

Rate Cv — AI skill for Claude Code

AI community

Claude Code skill that scores and critiques CVs against a weighted engineering-leadership rubric, with ATS and AI-slop checks and optional job-description matching.

How to install Rate Cv

This entry records only its repository, not the path inside it, so there is no exact command to give. Open maccydee/rate-cv and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Rate Cv does

Claude Code skill that scores and critiques CVs against a weighted engineering-leadership rubric, with ATS and AI-slop checks and optional job-description matching.

Alternatives in AI

  • Gtm Engineer Skills — Claude Code skill for improving website AEO (AI Engine Optimization) and GEO (Generative Engine Optimization) 761 ★
  • Sloptrim — A local detector for AI-writing patterns 196 ★
  • Describe Component — Write a six-section AI-optimised prose description of one component for Figma MCP and LLMs 194 ★

README

rate-cv

[![Claude skill](https://img.shields.io/badge/Claude-skill-8A2BE2.svg)](https://docs.claude.com/en/docs/claude-code/skills) [![License: MIT](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)

A Claude Code [skill](https://docs.claude.com/en/docs/claude-code/skills) that rates and critiques a CV/resume: it scores the document out of 100 against a weighted engineering-leadership rubric, checks ATS parse-ability, quantified impact, and AI-slop, and can match the CV to a specific job description with a keyword gap list.

The score is a sanity check, not truth. Real ATS platforms don't emit a score a recruiter sees, and commercial checkers disagree by ~20 points on the same file. So the skill leads with an auditable per-category breakdown and specific, coaching-framed fixes, not just a number.

What it does

  • Standalone quality (no job description): how strong the CV is for engineering-leadership roles, and what would raise it.
  • Targeted match (job description supplied): the above, plus a separate ATS keyword match and qualification-fit read against that role.
  • Mechanical signals from scripts/cv_signals.py: ATS parse-ability (format, single-flow vs tables/columns, section headers, dated roles, contact block), quantification density, action-verb usage, em-dash count, AI-slop terms, and JD keyword overlap.
  • Writing quality via an optional natural-writing linter pass, if present at ~/.claude/skills/natural-writing.
  • Judgement categories (leadership signals, impact specificity, seniority, fit) scored against references/rubric.md, using the bullet formulas and ATS rules in references/craft.md.

It never coaches toward fabrication: if a job wants a keyword the candidate can't truthfully claim, it reports a gap rather than suggesting you invent it.

Install

git clone https://github.com/maccydee/rate-cv.git ~/.claude/skills/rate-cv

Then in Claude Code, run `/rate-cv` (or just ask it to "rate my CV"). Wor