resume-tailor
npx machina-cli add skill Mahashwetha/Job-search-email-system-claude/resume-tailor --openclawResume Tailor
Generate a per-company tailored resume using Gemini 2.5 Flash via resume_tailor.py.
How to run
For a single job, run resume_tailor.py with the job posting URL and company name as arguments.
For batch mode (all companies with status done and non-LinkedIn links in the tracker), run it with no arguments.
Output
Saved to the resume_adjusted/ folder configured in config.py, named resume_{CompanyName}.docx. The script skips a company if the file already exists — delete the file to regenerate it.
What it does
Fetches the job description from the URL, reads the base resume, and asks Gemini to suggest minimal tweaks — reordering skills, surfacing existing keywords. It never fabricates experience. Gemini's markdown formatting (**bold**) is stripped automatically before writing to the DOCX.
Site compatibility
Most standard job pages work. Workday works via JSON-LD extraction. Ashby and internal portals (like BPCE) may fail — in that case, skip the URL and tailor manually, or paste the job description text into a temp file and adjust the call.
LinkedIn URLs are skipped automatically in batch mode.
If you hit rate limits
The free Gemini tier occasionally returns 429 errors — the script retries automatically. If it keeps failing, wait a minute and re-run. Use gemini-2.5-flash, not gemini-2.0-flash (that quota runs out faster).
Source
git clone https://github.com/Mahashwetha/Job-search-email-system-claude/blob/main/.claude/skills/resume-tailor/SKILL.mdView on GitHub Overview
Resume Tailor builds a per-company resume by reading the base resume and the job posting, then guiding Gemini 2.5 Flash to surface minimal tweaks such as reordering skills and surfacing keywords. It never fabricates experience and saves the adjusted resume as a DOCX in the resume_adjusted/ folder.
How This Skill Works
For a single job, run resume_tailor.py with the job posting URL and the company name as arguments. In batch mode, run it with no arguments to process all companies with status 'done' and non-LinkedIn links. Gemini suggests tweaks and strips markdown bold before writing the final DOCX to resume_adjusted/resume_{CompanyName}.docx.
When to Use It
- You have a specific company and want a tailored resume.
- You want a resume adjusted to a particular job posting URL or text.
- You need minimal tweaks (reorder skills, surface keywords) rather than a full rewrite.
- You run batch tailoring for multiple companies with a tracker (status 'done').
- You want a generated DOCX named per company and saved in the resume_adjusted folder.
Quick Start
- Step 1: Run resume_tailor.py with the job posting URL and the target company name as arguments.
- Step 2: Review the tweaks Gemini suggests; the output is written to resume_adjusted/resume_{CompanyName}.docx with bold stripped.
- Step 3: Open the DOCX, verify changes, and use the tailored resume for applications.
Best Practices
- Provide the job posting URL and company name when running.
- Use an up-to-date base resume as input for best results.
- Review Gemini's suggested tweaks and align keywords with the job description.
- Do not rely on Gemini to fabricate any new experience; it only tweaks and surfaces keywords.
- If a site blocks access or you hit rate limits, retry after a pause or delete the existing file to regenerate.
Example Use Cases
- Tailor a resume for Google's software engineer posting using a public JD URL.
- Adapt a resume for a finance analyst role posted by a specific company.
- Generate a resume for a customer success role at a named company.
- Tailor for an internal BPCE role when a suitable job URL is available.
- Run batch tailoring for multiple companies with status 'done' in the tracker.
