A structured workflow framework that runs locally to orchestrate AI agent tools for the end-to-end job search process. It requires the Claude Code CLI and Anthropic Claude models to operate, utilizing a drafter-reviewer pipeline for generating tailored application materials.
Project overview
It pairs a drafter-reviewer AI agent pipeline with an ATS simulation that extracts text from compiled CV PDFs to verify contact details, reading order, and keyword coverage.
Project type
AI Agent · AI Coding
Use cases
Coding & Development
Deployment
Refer to project documentation
License
MIT
Best for
Developers and technical job seekers who already use or are willing to set up the Claude Code CLI and want a structured AI workflow for customizing application materials.
Key capabilities
Drafts tailored CVs and cover letters in LaTeX, critiques them with a reviewer agent, and revises them to meet exact page constraints.
Extracts the text layer from compiled CV PDFs to verify contact details, reading order, and keyword coverage like an ATS parser.
Investigates local job boards and scaffolds new search skills with a standard CLI contract.
Limitations and risks
This is an independent project that is not affiliated with or endorsed by Anthropic.
Agentic defenses within the workflow operate at the instruction level rather than inside a sandbox; users should review fetched content before sending it.
Getting started
Installation involves cloning the repository and installing CLI tools via prerequisites that include Claude Code, Python, Bun, and LaTeX. Users run the 'claude' command and execute the /setup instruction to configure the environment.