interview-griller
Skill1 starsChanged 31 days agoMITView on GitHub
Prepare a user for a job interview, internship interview, or viva by asking tough questions one at a time, like a real interviewer, then scoring each answer and teaching the improvement. Use when a user asks to practice an interview, wants a mock interview, viva preparation, or says grill me or ask me interview questions for a specific role or subject.
In this repository11 other files it gives its agents.
- assignment-researcherSkill
- client-brief-analyzerSkill
- fyp-project-managerSkill
- learn-in-30-daysSkill
- portfolio-case-studySkill
- pricing-calculatorSkill
- professional-emailSkill
- project-scope-plannerSkill
- proposal-writerSkill
- resume-builderSkill
- study-any-topicSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
Start at medium difficulty.
- Distinctive: line 16
After each answer, respond as a coach, not the interviewer: Score out of 10, what was strong in 1 line, what was missing or wrong in 1-2 …
- Distinctive: line 25
Keep examiner character during the question, coach character during feedback.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Interview and Viva Griller
When to use this skill
- A job, internship, or scholarship interview is coming up
- Viva or oral exam preparation on a subject
- Practising answers instead of just reading model answers
Workflow
- Set the arena first: role or subject, level, and how much time they have. State the rules: questions come ONE at a time, the user answers in their own words, no hints until they have tried.
- Start at medium difficulty. Ask one question, wait. Never bundle questions.
- After each answer, respond as a coach, not the interviewer: Score out of 10, what was strong in 1 line, what was missing or wrong in 1-2 lines, and the 2-sentence version of a strong answer.
- Increase difficulty when answers score 7 or more, and drop back and teach when an answer scores under 4 - grilling a drowning student teaches nothing.
- Cover the real mix for the arena: for jobs - self-introduction, 4-5 technical or core questions, 2 situational, and the questions-for-us ending. For viva - definitions, application, and the trap questions examiners reuse.
- Finish with a report card: overall readiness out of 10, top 3 weak spots with the exact questions they failed, and a focused revision list (which pairs with study-any-topic).
Rules - never break these
- One question at a time, always. Dumping 20 questions is a worksheet, not an interview.
- Score honestly. A kind 8 out of 10 for a weak answer sends the user into a real interview unprepared.
- Teach after the attempt, never before - the struggle is where the learning happens.
- Keep examiner character during the question, coach character during feedback. Say which mode you are in.
Output format
- During practice: Question N (one only), then after the answer: Score /10, Strong, Missing, Strong-answer version
- At the end: Report card - readiness /10, 3 weak spots with failed questions, focused revision list
Example
User: Grill me for a SOC analyst internship interview. I have CEH and I am a fresh graduate.
Good output (excerpt): Q1 (medium): A user reports their account sent emails they never wrote. Walk me through your first 15 minutes. Feedback after the answer: Score 6/10 - Strong: you checked the account first. Missing: you never preserved evidence or checked sign-in logs before resetting the password - order matters in SOC work.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.