Interview Driven Learn

Interview-driven is all you need. Drives end-to-end tech learning with interview standards. Activated when the user submits study notes, project summaries, o...

Security Scan

The skill's declared purpose (turning study notes into interview-ready material and maintaining two local reference files) matches its instructions — no external installs or credentials — but it will persist user-submitted content to local files automatically, which can inadvertently store sensitive data.

Purpose & Capability

Name/description (interview-driven learning, question bank, knowledge base) align with the actual instructions: generate explanations, questions, STAR stories, analogies, and append results to references/question-bank.md and references/knowledge-base.md.

Instruction Scope

Instructions are explicit and limited to processing user learning inputs and appending structured entries to the two local reference files. There is no code, no network endpoints, and no requests for environment variables. However the skill instructs the agent to append user-provided content verbatim into persistent files, which could capture PII, proprietary code, or secrets if users paste them.

Install Mechanism

Instruction-only skill with no install spec, no binaries, and no downloads — lowest install risk.

Credentials

No environment variables, credentials, or external config paths are requested. The only resource accessed is the skill's own references/ directory for appending files, which is proportionate to the stated purpose.

Persistence & Privilege

Skill writes persistent data to its own references/knowledge-base.md and references/question-bank.md. It does not request always:true or system-wide changes, but persistent storage of arbitrary user inputs can be a privacy risk and may surprise users if triggered automatically.

Assessment

This skill is coherent with its stated purpose and doesn't require credentials or external installs. Before enabling or using it: (1) be cautious about pasting sensitive or proprietary text — the skill appends inputs verbatim to persistent files; (2) review or periodically purge references/knowledge-base.md and references/question-bank.md if they may contain secrets; (3) consider disabling automatic activation or requiring explicit confirmation before the skill runs to avoid accidental storage; and (4) verify file permissions for the references/ folder so only trusted users or processes can read them.

Interview Prep

Start from the end: turn every learning session directly into interview readiness.

Core Files

• Knowledge Base: references/knowledge-base.md — appended with each new topic, recording the theme + learning timestamp
• Question Bank: references/question-bank.md — all interview questions aggregated by topic for easy self-review

---

Input

Any learning content submitted by the user: study notes, technical concepts, project descriptions, etc.

Output: Five-Step Process

For every input, execute the following five steps:

---

Step 1 - Feynman Test (ELI5 + Professional)

Describe the concept in two ways:

• ELI5: As if explaining to a 10-year-old
• Professional: Complete, rigorous, no key details omitted

Purpose: Verify true understanding, not rote memorization.

---

Step 2 - Interview Question Generation

Generate 5-8 high-frequency interview questions in three categories:

• Fundamentals (what / differences / principles)
• Deep Dive (why / how / tradeoffs)
• Applied (examples / scenario-based)

Each question includes:

• What it tests
• Key answer points
• Follow-up direction if answered incorrectly

→ Also append to question-bank.md (aggregated by topic)

---

Step 3 - STAR Story Extraction

Break down the content into reusable STAR narratives:

• Situation: Background (technical scenario / business constraints)
• Task: Goal (what you needed to solve)
• Action: What you specifically did
• Result: Quantified outcomes + lessons learned

Best for: project experiences, problem-solving stories, team collaboration.

---

Step 4 - Analogical Learning (One to Three)

• Same-level analogy: What is this like in everyday life? What else works this way?
• Deeper analogy: What is one level below this? What's the underlying principle?
• Transfer analogy: Where else can this approach be applied?

Purpose: Build a knowledge network, not isolated facts.

---

Step 5 - Weakness Diagnosis + Knowledge Archive

Proactively uncover vulnerabilities:

• Where will interviewers probe until you can't answer?
• What do you think is important but actually isn't?
• What classic pitfalls remain unfilled? (edge cases, concurrency, distributed tradeoffs)

→ Append to knowledge-base.md with format:

## [Topic]
- Learned at: YYYY-MM-DD HH:mm
- Core takeaway: one-sentence summary
- Weak spots to reinforce: [spot 1, spot 2, ...]

---

Output Format Template

## 📚 Topic: [User's Input Topic]

---

### 1. Feynman Test

**ELI5:**
> [One-sentence version]

**Professional:**
> [Full description]

---

### 2. Interview Questions

| # | Question | Tests | Key Points |
|---|----------|-------|------------|
| Q1 |          |       |            |

**Follow-up traps:** ...

---

### 3. STAR Story

- **S**: [Background]
- **T**: [Goal]
- **A**: [Action]
- **R**: [Result + Reflection]

---

### 4. Analogical Learning

- 🔗 **Same-level**: ...
- 🔬 **Deeper**: ...
- 🚀 **Transfer**: ...

---

### 5. Weakness Diagnosis

⚠️ Likely follow-up pressure points:
1. ...
2. ...

---

*Synced to Knowledge Base & Question Bank*

---

File Structure

interview-prep/
├── SKILL.md
└── references/
    ├── knowledge-base.md   # Learning timeline
    └── question-bank.md    # Interview questions by topic

---

Trigger Words

When the user says/submits:

• "I learned XXX today"
• "Help me prepare for an interview"
• "Generate interview questions from these notes"
• "What interview questions can come from this concept"
• "What questions can this project be asked"

→ Activate this skill and run the five-step process, updating both documents.