← ClaudeAtlas

mock-interviewlisted

Generates a personalized mock-interview prep pack — persona-categorized question banks with grounded answer guidance and rationale, a fit-gap map, company-values alignment, reverse questions, and optional live drill sessions where Claude plays the interviewer. Use this whenever someone is preparing for a job interview, even if they don't say "mock interview" — triggers include "prep me for my interview", "interview questions for this role", "help me practice for my interview at X", "how should I answer questions about my background", an uploaded job description and/or CV in an interview context, or any mention of an upcoming recruiter screen, hiring-manager round, panel, or final round. Also use when tailoring prep to a specific interviewer or company. Candidate-side preparation only — not for employers screening candidates, and not for deciding whether to apply or whether a role is worth pursuing (that is evaluation, not rehearsal).
wavemanco/tuesday · ★ 0 · Testing & QA · score 72
Install: claude install-skill wavemanco/tuesday
# Mock Interview Build a prep pack a candidate can actually rehearse from. Two axes organize everything: **personas** describe how the interview will feel (who is asking, in what style), and **question types** describe what is being scored. Every question in the pack carries answer guidance grounded in the candidate's real material plus a rationale explaining why it's included and why it's categorized that way. Everything in the pack is either traceable to a provided document, traceable to research, or explicitly labeled as generic. **Started by its own command, `/mock-interview`, this skill keeps the turn:** `references/interaction-grammar.md` §6 says what that rules out — read it before reaching for any other skill. ## Before sending Copy this list and check the response against it. Each item points to its rule and restates none; rules 1, 5 and 7 are in `references/output-discipline.md`. ```text 1. A complete pack in the same message as any question (Step 1) 2. The pack's sections per the skeleton and conditionality table (Step 4) 3. The saved file stops at the upgrade menu; nothing of the message is in it (Step 4) 4. Questions within question_budget, counted across the message (rule 1) 5. The closing run in closing_order (Output discipline) 6. Watch-out line last in the message, exact, never in the pack file (rule 7) 7. Sibling named at most sibling_mentions times (rule 5) ``` ## Workflow ### Step 0 — Intake and classification Inventory every provided input agains