AgentSkillsCN

prep-for

根据价值映射、阶段 walkthrough 与针对性练习建议,制定专属于贵公司的面试备考计划。

SKILL.md
--- frontmatter
name: prep-for
description: "Build a company-specific interview preparation plan with value mapping, stage walkthroughs, and targeted practice recommendations"
disable-model-invocation: true
allowed-tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion, WebFetch
argument-hint: "<company: amazon|google|meta|custom> [--role <target role>] [--level <level>]"

Prep For

Generate a tailored preparation plan for a specific company's interview process, mapping the user's arsenal to company values and identifying gaps.

Steps

1. Load Configuration

Read .claude/what-did-you-do.local.md and .claude/i-did-a-thing.local.md. If either is missing, direct to setup.

2. Load Company Profile

If the user provided a company argument, load from <plugin_root>/companies/<company>.yaml (built-in) or <things_path>/interview-prep/companies/<company>.yaml (custom).

If no company specified, use AskUserQuestion:

Which company are you preparing for?

  • Amazon
  • Google
  • Meta
  • Custom — I have a job listing or company info to share

If "Custom", ask the user to paste a job listing or company information. Use it to build an ad-hoc company profile, referencing references/prep-strategy.md for structure.

3. Determine Target Role and Level

If --role and --level were provided, use them. Otherwise, check the user's profile for target_roles. If still unclear, ask:

What role and level are you targeting? (Free text — e.g., "Senior Software Engineer" or "Staff Engineer, L6")

Map the stated level to the company's level_expectations if available.

4. Load Arsenal and Session History

Read all arsenal files and session history. This powers the gap analysis.

5. Map Arsenal to Company Values

For each company value:

  1. Identify the value's interview_signals and question_themes
  2. Search arsenal files for entries that demonstrate these signals, considering all evidence types
  3. Search session history for relevant question answers
  4. Classify each value as:
    • Strong: Multiple arsenal entries with high relevance + strong session scores, with diverse evidence types
    • Partial: Some evidence but gaps in depth, variety, or evidence type diversity
    • Gap: Little or no evidence in arsenal or sessions

When matching, consider which evidence types are most relevant to each value:

  • Values about learning, adaptability, resilience → lesson entries
  • Values about technical depth, expertise → expertise entries
  • Values about judgment, ownership → decision entries
  • Values about influence, leadership, mentorship → influence entries
  • Values about vision, innovation, curiosity → insight entries
  • Values about delivery, results, bias for action → accomplishment entries

Present the mapping:

[Company] Values Alignment

ValueStatusEvidenceTypesGap
<value>Strong/Partial/Gap<best arsenal entry><evidence types present><what's missing>

6. Stage-by-Stage Preparation Guide

For each stage in the company's interview_process:

Stage: [Stage Name] ([format], [duration] min)

What they're evaluating: <evaluation_focus> Your interviewer type: <persona role>

Your readiness: <based on mock/practice session scores for this stage>

Preparation checklist:

  • Prepare [N] stories that demonstrate <values tested in this stage>
  • Practice with [persona] persona: /what-did-you-do:practice --persona <persona>
  • Run a mock for this stage: /what-did-you-do:mock <company> --stage <stage>

Key stories to prepare: (from arsenal)

  • <story from arsenal that maps to this stage's focus>

Anti-patterns to avoid: (from company values)

  • <anti-pattern from company profile>

7. Question Prediction

Based on company values and target level, predict likely questions:

Questions to Expect

High probability (directly maps to company values):

  • "<question>" — tests <value>
  • "<question>" — tests <value>

Medium probability (common at this level):

  • "<question>" — tests <skill>

Curveball potential (bar raiser / unique to this company):

  • "<question>" — tests <unusual skill or value>

8. Build Preparation Timeline

Ask the user:

When is your interview?

  • This week
  • Next week
  • 2-4 weeks out
  • 1+ month out
  • Not scheduled yet

Generate a preparation plan scaled to the timeline:

Preparation Plan: [Company] [Role] — [Timeline]

Week 1:

  • Day 1-2: Log any unlogged entries that map to [company] values — consider what evidence types you're missing (e.g., "For 'Learn and Be Curious,' log an expertise or lesson entry, not just an accomplishment")
  • Day 3-4: Practice [weakest category] with [persona] (2-3 questions/day)
  • Day 5: Mock [first stage] round

Week 2: (if applicable)

  • Continue practice focus areas
  • Mock [next stage] round
  • Review progress and adjust

[etc.]

9. Write Preparation Plan

Write the full plan to <things_path>/interview-prep/companies/<company>-prep-plan.md with frontmatter:

yaml
---
company: "<company>"
target_role: "<role>"
target_level: "<level>"
interview_date: "<date or null>"
created: <today>
values_coverage: <strong>/<total>
stage_readiness:
  <stage>: <ready|developing|gap>
arsenal_gaps:
  - "<skill or value>"
---

10. Offer Next Steps

Use AskUserQuestion:

Where do you want to start?

  • Log entries for my biggest gaps
  • Practice my weakest value area
  • Run a mock for the first stage
  • Review my full readiness
  • Done for now — I'll follow the plan