agentleFS
leoleolroleoleolro/career-os/.claude/skills/tailor/SKILL.md

tailor

Skill0 starsChanged 56 days agoMITView on GitHub

Produce a job-specific resume + cover letter (PDF/DOCX) derived from the master resume, for a job URL or an existing jobs/<slug> folder. Use when the candidate says "tailor", "make a resume for", "apply to this", "cover letter for", or pastes a job link.

10 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If a jobs/<slug>/jd.md exists, read it.

    line 9
  2. Very specific:

    Create jobs/<company-jobid-short-title>/ and write jd.md: header (title, company, job id, source URL, apply URL, location, posted/closing …

    line 10
  3. Distinctive:

    Never state salary figures as fact — give a band with "verify" or say unknown.

    line 41
  4. Distinctive:

    Check the printed page count and text-extraction line.

    line 51

“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/tailor <job url | jobs/<slug>>

1. Get the JD (verify it is open)

  • If a jobs/<slug>/jd.md exists, read it. Otherwise fetch the posting. Do not trust WebFetch on ATS sites — Phenom (BCG), Workday, Oracle ORC serve a "has been filled" decoy to plain fetches. Use the Browser pane (navigate → wait 3s → get_page_text) or the matching adapter: node -e 'import("./sources/index.mjs").then(async m=>{const s=m.getSource("phenom");console.log(JSON.stringify(await s.details({sourceId:"bcg-57792",url:"<url>"})))})'.
  • Create jobs/<company-jobid-short-title>/ and write jd.md: header (title, company, job id, source URL, apply URL, location, posted/closing dates, status + how verified), "What you'll do", "What you'll bring" (verbatim requirement bullets), and Fit notes: each requirement → the candidate's evidence from profile/master-resume.md, or "GAP".

2. Inputs (read all)

profile/master-resume.md (only source of facts), profile/certifications.md (what is obtained vs in progress), profile/preferences.json, profile/story-bank.md, profile/achievements.md (recent work not yet in master).

3. Write resume.md (ATS rules)

  • No em dashes (—) or en dashes (–) anywhere in resumes or cover letters, ever (the candidate, 2026-09-16: they read as AI-written). Use commas, colons, full stops, "to" for ranges, and "|" as the header/certs separator. Run grep -n '—\|–' jobs/<slug>/*.md before rendering; it must return nothing.
  • Each role gets its own heading, even internships (never combine two employers under one heading).
  • Certifications: list obtained ones; an exam that is booked may be listed as "(exam booked)"; the candidate's call (2026-09-16) is that an exam scheduled within the week of applying is listed as obtained, and the resume is regenerated the same day if the date slips.
  • Header: name, city, phone, email, LinkedIn as plain text (not a link); add github.com/<you> as plain text on engineering/AI resumes only, never on consulting ones. No photo, no columns, no tables, no icons.
  • One page for MBB (McKinsey/BCG/Bain) and graduate programs — render with --compact; two pages max for Australian industry / Big 4 lateral roles. Check the printed page count; cut, don't shrink below --compact. Headline line under the name mirrors the JD's title vocabulary truthfully (e.g. "Technology Consultant | Oracle EPM & Applied AI").
  • Summary/Profile: omit for MBB and graduate programs (not the convention — evidence only); for Australian industry / Seek / Big 4 roles keep a 3-line keyword-dense summary, no first person.
  • Skills block first for technical roles; Experience first for consulting roles.
  • Bullets: verb + what + tool + outcome/number. Reorder and reword master bullets to mirror JD language; never add a claim not in master. If the JD wants a number the candidate doesn't have, ask them — don't invent.
  • Certifications: list obtained ones with year; in-progress ones as "in progress (exam <date>)" — see certifications.md. Never present an unpassed exam as passed.
  • Consulting roles: include Leadership & Community (society leadership, volunteering). Graduate roles (BCG X): include extra-curricular/interests — ask the candidate if master-resume.md still says TODO.
  • Australian English. Client names: generalise ("ASX200 mining client") unless the candidate has said they can name them.

4. Write cover-letter.md

≤ 1 page, 4 paragraphs: (1) role + why this firm specifically (cite something real from the JD/company), (2) two proof points matching their top requirements, (3) the AI angle — what the candidate builds outside work, (4) availability, location/relocation, close. No clichés ("I am writing to…"). Date, addressed to the recruiter named in the JD if any (BCG ANZ: recruitmentinANZ@bcg.com).

4b. Write brief.md (the advice layer shown on the dashboard's Job packs tab)

Sections, in this order, grounded in the JD, profile/career-context.md, profile/preferences.json → priority_model (tiers, interest, stay_go_factors) and reports/market-insights-<latest>.md:

  1. Title line: # Brief — <role> · <P1/P2/P3> · <rating> then Verdict: one sentence (apply / apply with eyes open / skip) and why.
  2. Does it suit you? For / Against / Net — honest, specific to the candidate.
  3. Priority reasoning — brand direction vs the current employer (step up / lateral / step down), interest, pay (stated or "verify"), your current employer stay-vs-go context (scandal/restructure) and the neutral interview line.
  4. How to position yourself — which stories lead (story-bank numbers), what to de-emphasise, how to handle the years line, what to prepare/build before the screen.
  5. Trajectory of this role — title ladder at that company, typical timing, exit options.
  6. The company, briefly — what is known; mark anything unverified and point to /research <company>.
  7. Process to expect — stages, assessments; mark "typical — verify".
  8. Risks and unknowns · 9. Pack — links to resume/cover letter/interview-prep/research/network files. Never state salary figures as fact — give a band with "verify" or say unknown.

5. Render

export PATH="$HOME/.local/lib/node/bin:$PATH"
cd "$CAREER_OS"   # your career-os folder
node tools/render.mjs jobs/<slug> --name Resume_LeoLong_<Company>
# graduate programs that want one PDF (BCG X): add the transcript
node tools/render.mjs jobs/<slug> --name Application_LeoLong_<Company> --merge inbox/UWA_transcript.pdf

Check the printed page count and text-extraction line. Open the PDF (Read tool) and eyeball page 1.

6. Hand over

List the output files, the 3 biggest changes vs the Seek resume and why, any GAPs left, and what the candidate must confirm before applying. Then add to the pipeline: node tools/track.mjs add <url> --status shortlisted --company … --title … --next "Review + apply". Do not apply — that is /apply, and the candidate clicks submit.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.