Job application tailoring: Multi-Step Agent Workflow for Claude Code, Codex or ChatGPT

Researches the company, maps every job requirement to evidence on your resume, rewrites the resume for the role with a line-by-line change log and no invented experience, writes a cover letter from verifiable company hooks, checks ATS keyword coverage and preps interview answers.

Prompt

Installs to .agents/skills/job-application-tailoring-multi-step-agent-workflow-for-claude-c/

mkdir -p .agents/skills/job-application-tailoring-multi-step-agent-workflow-for-claude-c && curl -sL "https://prompts-gpt.com/prompts/job-application-tailoring-multi-step-agent-workflow-for-claude-code-codex-or-chatgpt/raw" -o .agents/skills/job-application-tailoring-multi-step-agent-workflow-for-claude-c/SKILL.md
SKILL.md
Preview
---
name: job-application-tailoring-multi-step-agent-workflow-for-claude-c
description: "Researches the company, maps every job requirement to evidence on your resume, rewrites the resume for the role with a line-by-line change log and no invented experience, writes a cover letter from verifiable company hooks, checks ATS keyword coverage and preps interview answers."
---

# Job application tailoring: Multi-Step Agent Workflow for Claude Code, Codex or ChatGPT

Researches the company, maps every job requirement to evidence on your resume, rewrites the resume for the role with a line-by-line change log and no invented experience, writes a cover letter from verifiable company hooks, checks ATS keyword coverage and preps interview answers.

## Prompt

```
---
name: job-application-tailoring
description: Researches the company, maps every job requirement to evidence on your resume, rewrites the resume for the role with a line-by-line change log and no invented experience, writes a cover letter from verifiable company hooks, checks ATS keyword coverage and preps interview answers. Use when asked to run the "Job application tailoring" workflow.
---

# Job application tailoring

Researches the company, maps every job requirement to evidence on your resume, rewrites the resume for the role with a line-by-line change log and no invented experience, writes a cover letter from verifiable company hooks, checks ATS keyword coverage and preps interview answers.

## Inputs

Ask for any of these that the user hasn't given:

- `resume` (e.g. "Paste your current resume as plain text")
- `job_description` (e.g. "Paste the full job post, including requirements and 'nice to have'")
- `company` (e.g. "Linear (linear.app)")
- `seniority` (e.g. "Senior software engineer, individual contributor")
- `constraints` (e.g. "Remote, EU only. Keep the resume to 2 pages. No relocation.")
- `extra_facts` (e.g. "True things not on the resume: led the move to PostgreSQL 15 in 2024; mentor two junior engineers. (or 'none')")

## Steps

Run the steps in order. Refer to saved results by name. Don't skip a step that says STOP.

1. **Paste resume and job post** — Start: Paste `resume` and `job_description`, name the `company`, and add `seniority`, `constraints` and any `extra_facts` that are true but missing from the resume.
2. **Research the company** — Research `company` for the role in `job_description`. Find: the mission in their own words, main products and customers, business model, news from the last 12 months with dates (launches, funding, leadership changes, layoffs), how they write (tone of their blog, careers page and docs), stated values, and the team this role sits in if that is public. Cite a URL for every fact. If the company name is ambiguous or you can't find the right one, say so and list what you found. End with 3 cover letter hooks: each a specific, recent, verifiable fact about the company that connects to something in `resume`. Save the result as `company_brief`.
3. **Map requirements to evidence** — From `job_description`, list every requirement and mark it must-have or nice-to-have using the post's own signals ('required', 'you have', 'bonus', 'nice to have'). For each, quote the matching line from `resume` or `extra_facts` and rate it: strong (direct, recent, with a result), weak (related, old, or no result) or missing.
Then list the ATS keywords: exact phrases from the post for skills, tools, certifications and the job title, with how many times each appears.
Finish with a fit summary: counts of strong, weak and missing must-haves, and the 3 biggest gaps. If the fit is poor for `seniority`, say so plainly. Save the result as `gaps`.
4. **Tailor the resume** — Rewrite `resume` for this job using `gaps`.
Hard rules: do not add employers, titles, dates, degrees, certifications, tools, numbers or responsibilities that are not in `resume` or `extra_facts`. Missing requirements stay missing.
You may: reorder sections and bullets, cut irrelevant bullets, merge bullets, rewrite wording, and use the job post's exact term where the resume shows the same thing in other words ('Postgres' → 'PostgreSQL'). Start each bullet with a verb and keep any result the original states. Add a 2–3 line summary aimed at the role. Respect `constraints`.
Return: (0) two lines at the top: 'Target: <the seniority>' and 'Constraints: <the constraints, in one line>'; (1) the full tailored resume as plain text with standard headings (Summary, Experience, Skills, Education); (2) a change log table: Section | Before (exact original text, or 'new') | After | Why (the requirement it serves). Mark any row that uses extra facts with '(from extra facts)'. Save the result as `resume_tailored`.
5. **Recruiter critique** — Review loop (max 3 passes): Act as a recruiter screening for the role named on the draft's 'Target:' line.
1. In the top third of the resume, a reader can see the target role and the 2–3 strongest matching qualifications.
2. Every bullet starts with a verb and states scope or a result; none say 'responsible for'.
3. Check the change log row by row: an 'After' must not contain any number, tool, employer, title, date or claim that is absent from its 'Before', unless the row is marked '(from extra facts)'. Fail any such row; the fix is to revert it to the 'Before' wording.
4. No keyword appears more than 3 times in the resume.
5. Plain text with standard headings; no tables or columns in the resume itself (the change log is separate).
6. The resume meets every item on the draft's 'Constraints:' line (for example, page length).
Feedback names the exact line and the fix. Keep the 'Target:' and 'Constraints:' lines and the change log in the revised draft. Save the result as `resume_tailored`.
6. **Write the cover letter** — Write a cover letter for the role in `job_description` at `company`, using `company_brief`, `gaps` and `resume_tailored`. 250–350 words in four short paragraphs: (1) why this company, using one hook from the research with a specific detail; (2–3) the top 3 must-haves, each matched to one piece of resume evidence and its result; (4) if a weak or missing requirement is likely to come up, one honest sentence about it, then a plain close. Match the tone of the company's own writing as described in the research. Use only facts from the resume and `extra_facts`. Do not write 'I am writing to apply', 'I believe I would be a great fit', 'passionate', 'dynamic' or 'leverage'. Save the result as `cover_letter`.
7. **ATS keyword check** — Compare the ATS keywords in `gaps` with `resume` (before) and `resume_tailored` (after).
1. Table: Keyword | Times in post | Before (yes/no) | After (yes/no, and where) | Note.
2. Coverage before and after, as a percentage of must-have keywords and of all keywords.
3. Keywords still missing, split into 'you may have this but it isn't on the resume: confirm before adding' and 'genuine gap'. Do not add them yourself.
4. Format checks on the text, each marked pass or fix: standard section headings, one consistent date format, contact details as plain text at the top, no key information that would sit in a table, column, text box, header or footer.
5. One line of export advice: save as .docx or a text-based PDF (not a scanned image), and keep the plain-text layout when you paste it into a template. Save the result as `ats`.
8. **Interview prep** — Using `gaps`, `company_brief` and `resume_tailored`:
1. 8 likely interview questions: 4 on must-haves, 2 probing weak or missing items, 2 about the company or its recent news. For each, name the resume story to use (role and bullet) and give a 3-line STAR outline built only from facts on the resume.
2. 4 questions to ask the interviewer, each tied to something in the research.
3. The one gap to prepare for, with a suggested answer that is honest and does not overclaim. Save the result as `prep`.
9. **Check every change** — STOP and ask the user to approve before continuing: Read the change log in `resume_tailored` line by line: only you know what's true. Then review `cover_letter`, `ats` and `prep`. Approve or edit before sending anything.
10. **Save the application pack** — Finish: Return `resume_tailored`, `cover_letter`, `ats`, `prep` and `company_brief`.

## Done when

Every step has run, each approval was given, and you've reported what was produced and where it was saved.
```

## Usage Notes

Run it in Claude Code or Codex: save as .claude/skills/job-application-tailoring/SKILL.md (or .agents/skills/ for Codex) and say "run the Job application tailoring workflow"; parallel steps run as subagents. Or paste it into ChatGPT or Claude and answer its questions. It asks for: resume, job_description, company, seniority, constraints, extra_facts. Build or run it step by step at prompts-gpt.com/workflows/job-application-tailoring.

Example output

At a glanceATS keyword coverage before and after tailoringMaya Okafor's resume for Tallyfin Senior Backend Engineer
ItemBeforeAfter
Must-have keywords58%83%
All keywords47%71%

Example for: Maya Okafor, backend engineer (6 yrs), applying to Tallyfin (invoicing SaaS), Senior Backend Engineer, Remote EU, 2-page limit. Extra facts: led PostgreSQL 15 upgrade in 2024; mentors two junior engineers.

Steps 1–8 complete. Awaiting your approval (Step 9)

company_brief (summary)
  • Mission: "Make getting paid boring" (tallyfin.example/about)
  • News: Series B, €38M (14 Mar 2025, tallyfin.example/blog/series-b). Launched EU e-invoicing for Peppol (2 Sep 2025, tallyfin.example/changelog)
  • Tone: Plain, dry, short sentences, no hype
  • Cover letter hooks:

    1. Peppol launch ↔ Maya's EDI integration work at Brightcart
    2. Series B scaling ↔ her 4× throughput queue redesign
    3. "Boring reliability" value ↔ her on-call incident reduction
gaps (fit summary)
Must-havesStrongWeakMissing
751 (Kafka: 2019, no result)1 (Go)

Biggest gaps: Go, Kafka recency, no public-facing API design shown.
Fit: Good for senior IC.

resume_tailored: change log excerpt
SectionBeforeAfterWhy
SummarynewBackend engineer, 6 years building payment and invoicing services in Python and PostgreSQL…Top-third match
Brightcart"Worked on Postgres upgrade""Led upgrade to PostgreSQL 15 with zero downtime" (from extra facts)PostgreSQL (must-have)
Brightcart"Rebuilt job queue, 4x throughput""Redesigned job queue on Redis, raising throughput 4x"Scalability

Recruiter critique: Ran 2 passes. Pass 1 failed one row because "zero downtime" was not in the source. That wording was reverted to "Led upgrade to PostgreSQL 15". Pass 2 passed all checks, and the resume fits in 2 pages.

cover_letter

Four paragraphs, 291 words. It opens on the Peppol launch hook. It has one honest line on Go: "I haven't shipped Go in production; my last two services were Python."

ats
CoverageBeforeAfter
Must-have keywords58%83%
All keywords47%71%

Confirm before adding: "Terraform" (you mentioned infra work. Did you write Terraform?)
Genuine gap: Go
Format checks: All passed. Fixed one date-format inconsistency (Mar 2021 → 03/2021, applied throughout).
Export: Save as .docx or a text-based PDF.

prep
  • 8 questions. Example: "Tell us about scaling a system under load." Use the Brightcart queue bullet. Situation: 2am backlogs. Task: own the fix. Result: 4× throughput.
  • 4 questions to ask. Example: "How did Peppol rollout change on-call load?"
  • Gap to prepare for: Go. Suggested answer: "I read Go codebases and have done small internal tools, not production services. I'd expect a few weeks to be productive."

⏸ STOP: Step 9 approval needed

  1. Read every change log row. Only you know what's true. Confirm the "(from extra facts)" rows in particular.
  2. Answer the Terraform question.
  3. Check that the Go answer in prep is accurate.

Reply "approve" or send edits. Once you approve, I'll save the full application pack.

Written by AI from this prompt. Your results will vary.

Run it in Claude Code or Codex: save as .claude/skills/job-application-tailoring/SKILL.md (or .agents/skills/ for Codex) and say "run the Job application tailoring workflow"; parallel steps run as subagents. Or paste it into ChatGPT or Claude and answer its questions. It asks for: resume, job_description, company, seniority, constraints, extra_facts. Build or run it step by step at prompts-gpt.com/workflows/job-application-tailoring.

Make it repeatable

Keep using this prompt without pasting it in every time.

Prompt details

By
Prompts-GPT editors
Published
How prompts are written and checked

Did this prompt work?

Keep exploring

More in AI Marketing