/interview-prep

Usage

/interview-prep <company name + role>
/interview-prep <company name + role> recruiter-screen

How It Works

┌─────────────────────────────────────────────────────────┐
│                    INTERVIEW PREP                        │
├─────────────────────────────────────────────────────────┤
│  Step 1: CONTEXT                                         │
│  - Read JD + candidate profile + stories                │
│  - Detect company → load matching framework              │
│    (e.g., amazon.md for LPs, mckinsey-bain-bcg.md)      │
│                                                          │
│  Step 2: PREP PACKAGE                                    │
│  - Likely questions with STAR frameworks                 │
│  - Power stories mapped to question types                │
│  - Smart questions to ask                                │
│  - Red flags + compensation notes                        │
│                                                          │
│  Step 3: FOLLOW-UP QUESTIONS                             │
│  - Scenario questions from JD responsibilities           │
│  - Deep-dive follow-ups for later rounds                 │
│  - Culture + motivation probes                           │
└─────────────────────────────────────────────────────────┘

Inputs

• Job description (provided by user)
• knowledge/profile.md — candidate context
• knowledge/stories/ — STAR story bank
• knowledge/frameworks/writing-framework.md — for structuring verbal answers
• knowledge/values/personal.md — for culture fit and motivation questions
• If available: output/[company]/company-brief.md — use company intel to tailor
• interview-type (if specified): recruiter-screen / technical / hiring-manager / final
• ${CLAUDE_PLUGIN_ROOT}/.claude/resources/ — detect the company or career path and load

the matching framework. Search companies/ first (e.g. amazon.md, google.md), then paths/ (e.g. consulting.md, pe.md, marketplace.md), then fall back to general.md.

Produce

1. Interview Format Prediction

Based on role and company: likely stages, who you'll meet, what each round focuses on.

2. Likely Questions (8-12)

For each:

Q: [question] Why they ask it: [what they're really evaluating] STAR Framework:

• Situation: [what context to set]
• Task: [your responsibility]
• Action: [specific steps you took]
• Result: [quantified outcome]

Best story: [which story from knowledge/stories/ and why]

3. Power Stories (3)

From knowledge/stories/, identify the 3 that answer the widest range:

• The story in ~2 minutes
• What it proves about you
• Which questions it can answer

4. Questions to Ask the Interviewer (8-10)

Smart questions that signal deep research, strategic thinking, and genuine curiosity.

5. Red Flags to Prepare For

Based on profile gaps — how to address proactively.

6. Anticipated Follow-Up Questions

First-Round Follow-Ups: Q: [question] What they're really testing: [underlying concern] What a strong answer looks like: [framework, not script]

Scenario Questions: Scenario: "Imagine you're 3 months in and..." What they want to see: [decision framework] Candidate's best angle: [which story to draw from]

Culture + Motivation Probes: Why this company, why now, why you — specific to this JD.

7. Compensation Notes

Expected range for role/level, when to bring it up, how to phrase it.

Output Format

## Interview Format: [Company] — [Role]
**Likely stages:** [stages]
**This round focuses on:** [focus]

---

## Top Questions

### Q1: [question]
**Why they ask:** [evaluation target]
**STAR:**
- **S:** [situation]
- **T:** [task]
- **A:** [action]
- **R:** [result — with numbers]
**Use story:** [story name]

...

## Power Stories
| Story | Proves | Answers |
|-------|--------|---------|
| [name] | [trait] | [question types] |

## Questions to Ask
1. [question] — *signals [what]*

## Red Flags
| Gap | How to Address |
|-----|----------------|
| [gap] | [proactive framing] |

## Follow-Up Questions to Prepare
...

## Compensation
Range: $X-Y | Bring up: [when] | Phrasing: [how]

If Something Goes Wrong

• No company framework found: Use general.md as fallback. Note: "Using general behavioral framework — tell me the company for more specific prep."
• No stories match a likely question: Flag it: "You don't have a prepared story for [topic]. Run /personalize stories to add one."
• No knowledge base at all: Stop and say: "Run /personalize first — I need your profile and stories to prep effectively."
• Candidate appears underqualified: Say so directly. Focus on reframing strengths, not hiding gaps.

After Output

• "Want to practice? /mock-interview for a simulated interview"
• "Track this in your pipeline? /pipeline-status add [company] [role]"

Rules

• Ground every story in actual stories from knowledge/stories/
• Be specific to THIS role and company
• Flag if candidate is underqualified and needs to reframe
• Note where founder experience helps vs needs framing
• Note where education/MBA angle is an asset
• Structure verbal answers using the candidate's writing framework