ワンクリックでManusで任意のスキルを実行

resume-tailor

Use whenever Alan Yu is applying to a specific job and needs his resume or cover letter tailored to that opportunity. Triggers on: pasting a job description, asking 'how should I tailor this for [role]', requesting a cover letter for a specific company, asking for fit analysis between Alan's background and a posted role, or producing a .docx resume targeted at a specific employer. Do NOT use for unrelated resume work or for other people's resumes.

Manusで実行

概要

インストールコマンド

npx skills add https://github.com/dtsong/my-claude-setup --skill resume-tailor

このコマンドをClaude Codeにコピー＆ペーストしてスキルをインストール

ソース

dtsong/my-claude-setup

スター5

フォーク0

更新日2026年5月26日 05:22

ファイルエクスプローラー

13 ファイル

SKILL.md

readonly

このリポジトリの他の Skills

同じリポジトリ

docx-to-pdf

dtsong/my-claude-setup

Use to convert a Word .docx file to PDF and/or verify its page count. Triggers on: converting docx to pdf, rendering a document, checking how many pages a docx produces, or asserting a page-count constraint (e.g. a resume must stay 2 pages). Wraps LibreOffice headless conversion.

2026-05-265

evaluate-diagram

dtsong/my-claude-setup

Use this skill when scoring or comparing a generated diagram against a human reference. Triggers on "score this diagram", "evaluate my diagram", "compare to reference", or "how accurate is this". Applies when both a generated diagram and a reference image exist and quality assessment is needed. Do NOT use for creating new diagrams (use generate-diagram) or plotting data (use generate-plot).

2026-04-065

generate-diagram

dtsong/my-claude-setup

Use this skill when creating a methodology diagram from research text. Triggers on "make a diagram", "visualize this methodology", "diagram this process", or "generate a figure from this paper". Applies to methodology descriptions, process flows, and research paper sections. Do NOT use for scoring existing diagrams (use evaluate-diagram) or plotting data from CSV/JSON (use generate-plot).

2026-04-065

generate-plot

dtsong/my-claude-setup

Use this skill when creating a statistical plot or chart from a data file. Triggers on "plot this data", "make a chart", "graph this CSV", or "visualize these results". Applies to CSV, JSON, or tabular data needing bar charts, scatter plots, line graphs, or similar visualizations. Do NOT use for methodology diagrams from text (use generate-diagram) or diagram scoring (use evaluate-diagram).

2026-04-065

verilator-simulation

dtsong/my-claude-setup

Use when planning or reviewing Verilator-based simulation workflows for SystemVerilog designs. Covers lint analysis, simulation setup, trace/waveform debugging, coverage-driven verification, and C++ co-simulation. Do not use for commercial EDA tools (use verification-methodology) or RTL design flow (use chip-design-flow).

2026-03-305

a11y-audit

dtsong/my-claude-setup

Use when auditing accessibility compliance of a feature or codebase. Covers WCAG 2.2 AA conformance, screen reader compatibility, keyboard navigation, focus management, color contrast, reduced motion, and ARIA usage. Do not use for general UX review (use journey-mapping) or component interaction specs (use interaction-design).

2026-03-265

ソース

dtsong

dtsong/my-claude-setup

GitHub リポジトリを開く Creator のリポジトリを見る

インストールコマンド

ダウンロード

Manusで実行

役立つ用途SOC

作家・著者芸術・デザイン・エンタメ・スポーツ・メディア職27-3043L4

name

resume-tailor

description

resume-tailor

A skill for producing job-specific tailored versions of Alan Yu's resume and a matching cover letter draft. Always grounded in his actual experience — never fabricates accomplishments or extrapolates beyond what's in base/alan-resume.md.

When to use

The user pastes a JD (job description) and wants a tailored application package. Common phrasings:

"Here's a JD, tailor my resume."
"How well does Alan fit this role?"
"Draft a cover letter for this [company] role."
"Build me a .docx version tailored to this posting."

If the user just wants the base resume unchanged, skip the tailoring workflow — point them to base/alan-resume.md and scripts/build.js directly.

Workflow

Follow these steps in order. Each builds on the prior.

1. Load context

Read in this order:

base/alan-resume.md — canonical source of truth for Alan's experience
base/alan-voice.md — framing rules, banned phrasings, verification status of metrics
The user-provided JD

If alan-voice.md flags an open verification question that touches the JD (e.g., MTTR metric, AI bullet specifics), surface that to the user before producing output.

2. Analyze fit

Produce a brief fit analysis (4-6 bullets max). Categorize each JD requirement as:

Direct hit — Alan has clear, quantified experience matching this
Adjacent — Alan has related experience that can be reframed honestly to fit
Gap — Alan does not have this experience; flag for the user; do NOT paper over

Never invent matches. If the JD requires Kubernetes and Alan has never touched it, say so.

3. Calibrate framing

Based on the JD's industry signals, decide whether default retail framing fits or whether to propose adjustments per alan-voice.md § "Industry calibration":

Retail / E-commerce / D2C → default (no changes needed)
Healthcare / Biotech → propose subtitle and bullet adjustments
Defense / Gov → propose subtitle and bullet adjustments; flag clearance gap if required
Hospitality / Multi-Site → propose subtitle adjustment to surface multi-site scope
TPM / Program Manager → propose PM-flavored subtitle and verb framing

Default: retail (Alan's authoritative positioning). For non-retail JDs, present the proposed calibration to the user before generating output — do not silently substitute. Only emphasize luxury retail when the JD is explicitly luxury.

4. Produce tailored content

Generate a JSON file at output/tailored-resume.json matching the schema in base/alan-resume.json (the structured form of the resume). Tailoring is allowed in these specific ways:

Allowed:

Rewriting the summary paragraph to lead with traits the JD emphasizes
Reordering bullets within a role to surface relevant work first
Lightly rephrasing bullets to weave in JD keywords WHERE THE UNDERLYING WORK MATCHES
Adjusting the subtitle line ("Retail Systems Manager · ..." → tuned to JD)
Including/excluding optional bullets when length needs adjustment

Not allowed:

Adding metrics, technologies, accomplishments, or scope not present in base
Changing employer names, dates, or core titles (the "(Americas Lead)" descriptor on Hermès stays exactly as in base; nothing else changes)
Inventing JD-keyword matches by twisting Alan's actual work into shapes it didn't take

5. Build the .docx

Run:

node scripts/build.js output/tailored-resume.json output/Alan_Yu_Resume_[CompanyName].docx

The build script consumes the JSON and emits a styled .docx matching Alan's established design (navy/blue color scheme, Arial, two-page layout).

Verify the 2-page constraint. Render to PDF and check the page count using the docx-to-pdf skill:

node ~/.claude/skills/docx-to-pdf/scripts/docx2pdf.js output/Alan_Yu_Resume_[CompanyName].docx /tmp

If pages > 2, compress lower-priority bullets (never the Hermès section) and rebuild. Do not ship a 3-page resume.

6. Draft cover letter

Follow templates/cover-letter.md. It encodes the structure, the verified-safe proof points, and the pre-return checklist. Read base/alan-voice.md § "Cover letter voice" first.

Write a 180–220 word letter to output/Cover_Letter_[CompanyName].md:

Opening (2-3 sentences): why this role + why Alan; name one concrete fact about the company
Middle (one proof point per paragraph): 2-3 verified accomplishments mapped to JD priorities
Close (1-2 sentences): specific interest + call to action; end with the San Diego line

Voice: confident, not boastful. First person. Active verbs. Alan's measured palette (do not silently upgrade to owned/drove). No "I am writing to apply for" filler. Em-dash cap 0–1 across resume AND letter combined.

7. Output summary

Produce a short "what I tailored and why" note (5-8 bullets) so the user can review changes before sending. Flag any gaps surfaced in step 2 explicitly.

8. Log the run

Append one row to output/applications.md (create it from the header in that file if missing). Use:

Date — today (YYYY-MM-DD)
Company / Role — from the JD
Status — Drafted (Alan updates as it progresses)
Follow-up — Date + 7 days, or — if not applying yet
Files — the generated .docx and cover letter filenames

One row per invocation. Never rewrite existing rows; only append.

Weekly portal scan

For sourcing roles across Alan's San Diego target list (output/target-companies.md):

Regenerate the checklist: node scripts/job-search-links.js → writes output/weekly-search.md (per-company LinkedIn / Indeed / careers search links, scoped to San Diego). No scraping; Alan opens the links.
For each promising posting, run the quick triage in templates/fit-triage.md (0–10 score, go/no-go).
For Strong hits (and chosen Maybes), run the full workflow above via /tailor <jd-path> or by pasting the JD.
Respect the ⚠️ clearance flags carried into the checklist — auto-skip roles requiring an active clearance (Alan has none).

This is a sourcing aid, not an applicant. It generates search links and scores fit; a human reviews and applies.

Hard constraints

These apply to every invocation, regardless of user request:

Truth before polish. If tailoring requires invention, stop and surface the gap.
Never modify base files. base/alan-resume.md, base/alan-resume.json, and base/alan-voice.md are read-only sources of truth. Updates to them happen through a separate workflow with Alan's confirmation.
Preserve dates and employers exactly. No exception.
Two-page maximum for the .docx output. If content overflows, compress lower-priority bullets, not the Hermès section.
Output to output/ directory. Never write to base/ or scripts/.
Em-dash hard cap: 0–1 em-dashes across the entire resume and cover letter. See alan-voice.md § "Em-dash restriction." Use commas, semicolons, or sentence restructuring instead. This is non-negotiable — em-dashes are a recognizable AI-generation signal and Alan has asked they be limited.
Preserve Alan's voice. Do not silently upgrade his verbs (e.g., "managed" to "owned"). If a stronger framing would meaningfully help for a senior role, propose it in the change summary and let Alan decide.

File structure

resume-tailor/
├── SKILL.md                    # this file
├── README.md                   # install + first-time setup
├── base/
│   ├── alan-resume.md          # human-readable canonical content
│   ├── alan-resume.json        # structured form (consumed by build.js)
│   └── alan-voice.md           # framing rules + verification status
├── scripts/
│   ├── build.js                # JSON → .docx
│   └── job-search-links.js     # target-companies.md → weekly search links
├── templates/
│   ├── cover-letter.md         # cover letter scaffold + checklist
│   └── fit-triage.md           # quick go/no-go scoring rubric
├── examples/
│   └── tailored-example.json   # one worked example for reference
└── output/                     # generated files land here (created on first run)

Updating the base

When Alan provides new information (e.g., confirms the MTTR metric, adds a new accomplishment, gets a title change), update base/alan-resume.json AND base/alan-resume.md. Then update base/alan-voice.md to remove the corresponding verification flag. The build script reads from JSON; markdown is for humans.

Dependencies

Node.js 18+
docx npm package: npm install -g docx