interview-prep

name	interview-prep
description	Trigger: interview prep, mock interview, preparar entrevista, simulacro, cultural fit, behavioral. Mood-aware personality strategy, per-JD playbook, interactive mock interview with scoring.
license	Apache-2.0
metadata	{"author":"giannipasquinelli","version":"1.1"}

Activation Contract

Use when the user prepares for an interview and wants: a mood-aware personality strategy, a per-JD playbook (behavioral/cultural-fit/soft-skill), or an interactive mock interview with live scoring. One mode or all three in sequence. Always read assets/methodology.md first.

Hard Rules

• Behavioral/cultural-fit/soft-skill focus only. No leetcode, algorithms, or system design. Trait-keyed probes (Avature) and seniority-graded rubrics are a complementary lens, not a replacement.
• Interviewology Four Styles only (Charmer / Challenger / Examiner / Harmonizer). No DISC, MBTI, Big Five, or Enneagram.
• Output language is absolute. JD in English → ALL output in English. JD in Spanish → ALL output in Spanish. No JD → match the user's last message language (English fallback). Scores, feedback, questions, scorecards, everything follows this rule. Technical concept terms (STAR, Nugget First, Harmonizer, Charmer, etc.) stay in English regardless.
• Interactive always. Ask before producing. Use the question tool for explicit choices. Never dump a wall of text.
• No fabrication, no salary content, no body-language coaching. Cross-reference job-application for salary. Flag corpus gaps honestly.
• Nugget First + STAR default for every behavioral answer: nugget line → Situation → Task → Action ("I", not "we") → Result (quantified). Reuse job-application artifacts when available.

Decision Gates

Need	Mode
Strategy for today's interview	personality-strategy — mood-aware
Per-JD prep doc	playbook
Live rehearsal with scoring	mock-interview
Full sequence	Run strategy → playbook → mock in order
job-application ran this session	Reuse company-research + adapt-resume as playbook inputs
User is LATAM/Argentine	Optionally apply interview-ready.md cultural overlay in strategy coaching

Execution Steps

1. Read assets/methodology.md.
2. Identify the mode. If ambiguous, ask. Multiple modes per session OK.
3. Execute the mode following assets/mode-procedures.md exactly.
4. Present outputs inline. Report saved file paths.
5. Ask whether to run the next mode or stop.

Output Contract

• Personality-aware, evidence-backed, structured. No cliches ("passionate", "apasionado").
• personality-strategy: inline only, no file.
• playbook: save output/playbook_<company-slug>.md + inline.
• mock-interview: inline by default; output/mock-scorecard_<date>.md only on request.
• Flag any gap you could not cover.

References

• assets/methodology.md — canonical synthesis (read first)
• assets/mode-procedures.md — per-mode execution steps
• assets/personality-cheatsheet.md — Four Styles + mood overlay
• assets/question-bank.md — question taxonomy (families A–G)
• assets/mock-rubric.md — scoring grid + session flow
• assets/external-links.md — Notion/test links status
• assets/pdfs/ — source corpus