cv-tailor

Files (2)

• LICENSE
• SKILL.md

SKILL.md (first lines)

# CV Tailor

**Three pillars of resume optimization**: Analyze keyword alignment against the target JD, rewrite experience bullets using the STAR method with quantified results, and run an ATS compatibility check — producing a highly targeted, high-pass-rate optimized resume.

## Quick Start

The user provides their resume (content or file) and the target JD. The agent then automatically completes the optimization following the workflow below:

```
User: Help me optimize my resume — I'm applying for this role [attaches JD + resume]
Agent: [Follows the SOP workflow and outputs optimization recommendations plus a rewritten resume]
```

## SOP Workflow

### Phase 1: Input Collection & Initial Analysis

**Goal**: Gather the user's resume and target JD; establish an optimization baseline.

**Steps**:

1. **Collect materials**:
   - Obtain the user's resume content (pasted text or file path)
   - Obtain the target JD (pasted text or role description)
   - If no JD is provided, ask about the target role direction (industry + position + level)

2. **Resume baseline parsing**:
   - Identify resume sections (education, work experience, projects, skills, etc.)
   - Count resume length, number of experience entries, and time span
   - Note the current resume format type (reverse-chronological / functional / hybrid)

3. **JD core element extraction**:
   - Job title and level
   - Core responsibilities (Top 5)
   - Hard requirements (must-haves)
   - Nice-to-haves
   - Key skill terms and industry jargon

**Output**: Resume status summary + JD element checklist

---

### Phase 2: JD Keyword Match Analysis

**Goal**: Systematically compare keyword coverage between the resume and JD to identify match gaps.

**Steps**:

1. **Categorized keyword extraction**:
   Extract three categories of keywords from the JD:

   | Category | Description | Examples |
   |----------|-------------|----------|
   | **Hard skill keywords** | Tech stack, tools, methodologies | Python, SQL, A/B testing, Scrum |
   | **Soft skill keywords** | Competency requirements | Cross-team collaboration, data-driven, project management |
   | **Industry/domain keywords** | Domain-specific terminology | DAU, conversion rate, user growth, SaaS |

2. **Match analysis**:
   Search each keyword in the resume and generate a match matrix:

   ```
   | Keyword | JD Priority | In Resume? | Location | Recommendation |
   |---------|-------------|------------|----------|----------------|
   | Python  | Required    | ✅ Yes     | Skills + Project 1 | Keep; add specific use-case context |
   | SQL     | Required    | ❌ No      | -        | Add; weave into project experience |
   ```

3. **Coverage scoring**:
   - Required keyword coverage = matched required keywords / total required keywords × 100%
   - Nice-to-have coverage = matched nice-to-have keywords / total nice-to-have keywords × 100%
   - **Benchmark**: Required keyword coverage ≥ 80% is passing, ≥ 90% is excellent

4. **Gap-fill recommendations**:
   - For each unmatched required keyword, recommend which section and entry to add it to
   - Provide specific integration approaches (add to skills section / embed in experience bullet / highlight in project outcomes)

**Output**: Keyword match matrix + coverage scores + gap-fill plan

---

### Phase 3: STAR Quantified Rewriting

**Goal**: Rewrite each experience entry using the STAR method, ensuring quantified data support.

**STAR Method Definition**:

| Element | Meaning | Checkpoint |
|---------|---------|------------|
| **S** - Situation | Context & background | When, what scenario, what scale |
| **T** - Task | Objective & responsibility | What was your role, what problem to solve |
| **A** - Action | Specific actions taken | What you did, what methods/tools you used |
| **R** - Result | Quantified outcomes | Data changes, efficiency gains, cost savings |

**Steps**:

1. **Diagnose existing entries**:
   Evaluate STAR completeness for each experience bullet:

   ```
   Original: "Responsible for user growth initiatives"

   Diagnosis:
   - S (Situation): ❌ Missing — no product or stage context
   - T (Task): ⚠️ Vague — "initiatives" is too generic
   - A (Action): ❌ Missing — no specific actions described
   - R (Result): ❌ Missing — no data whatsoever
   Score: 1/4 (severely lacking)
   ```

2. **Quantified rewriting**:
   After gathering additional details from the user, rewrite using the STAR structure:

   ```
   Rewritten: "During a user growth plateau for [Product Name] (DAU 500K+),
   led the design of a new-user activation funnel analysis framework (S+T),
   optimized 3 critical registration flow touchpoints + designed a 7-day retention incentive strategy (A),
   increasing new-user D1 retention from 32% to 45% and monthly active users by 18% within 3 months (R)"
   ```

From zebbern/claude-code-guide, MIT.

More from zebbern/claude-code-guide