Hr Interview Question Generator Copilot

HR Interview Question Generator

Purpose

Generate candidate-specific interview plans after resume screening.

The goal is to verify whether the candidate truly has the experience claimed in the resume and whether they fit the JD.

Inputs

Accept:

• JD text or JD file.
• Candidate resume text.
• Candidate detail report.
• Candidate names selected by the user.
• Specific interview focus requested by the user.

If candidate resumes are missing, ask for them or ask the user to run hr-resume-screening-copilot first.

Shared Resources

Use HR Agent Copilot shared files:

• ../references/interview-template.md
• ../references/scoring-rubric.md

If installed in a flattened environment, locate equivalent references/ paths near the skill root.

Optional LLM Wiki Augmentation

This skill declares its memory contract in scp.yml. Before the main business workflow, read ../references/llm-wiki-integration.md, run the resolve-config preflight, and apply the declared query flow when HR memory is enabled. If any runtime step fails, follow the shared fallback contract and complete the original workflow.

Workflow

When the user names a candidate without supplying the resume again, apply the shared Candidate Resolution flow before concluding that the resume is missing. Do not search Graph output or candidate directories as a fallback.

For each selected candidate:

1. Read the JD requirements.
2. Identify the candidate's strongest matching claims.
3. Identify the candidate's highest-risk or most ambiguous claims.
4. Turn those claims into targeted validation questions.
5. Prefer scenario and project-depth questions over textbook questions.
6. Provide reference answer points.
7. Provide follow-up probes.
8. Provide weak-answer and rejection signals.
9. End with a concise interview recommendation focus.

Question Types

Use a mix of:

• Resume-claim validation questions.
• Scenario design questions.
• Incident debugging questions.
• Trade-off questions.
• "What did you personally build?" questions.

Avoid relying only on:

• Textbook definitions.
• Generic "tell me about your project" questions.
• Questions unrelated to the JD.

Output Structure

Use this structure for each candidate:

# [Candidate Name] Interview Plan

## Candidate Positioning

...

## Interview Focus

1. ...

## Questions, Reference Answers, And Follow-Ups

### 1. [Question]

Why ask:

- ...

Reference answer points:

- ...

Follow-up probes:

- ...

Weak answer signals:

- ...

## Risk Signals

1. ...

## Post-Interview Scoring Guide

- Strong pass:
- Pass:
- Weak pass:
- No pass:

Reference Answer Rules

Reference answers should be answer points, not rigid scripts.

Good answers usually include:

• System boundary.
• Data model.
• State transitions.
• Failure modes.
• Idempotency and retry.
• Security and permission checks.
• Observability.
• Trade-offs.
• Concrete examples from the candidate's own work.

Risk Verification

Always include questions that verify:

• Whether the candidate personally built the claimed module.
• Whether the candidate can explain design trade-offs.
• Whether the candidate has handled failures in production.
• Whether keyword claims are shallow or deep.