FS

fenlili0108-source/interview-coach

Developer tools

This page was translated automatically. Technical details may differ from the original. View the original

82 stars Quality 54 Trend 19MIT

General AI Job Interview Assistant · A reusable Claude Code skill for interview prep, live rescue & post-interview review. Fill in your own profile.

Overview

General AI Job Interview Assistant — Suitable for any position and industry. All customized capabilities come from **; skill itself contains no personal information. This is a Claude Code: before the interview, it helps you conduct company research, JD analysis, Q&A prediction, and rhetorical question design. During the interview, you get on-the-spot scripts, and afterward, you record reviews and diagnose weak points. Ready to use out of the box, privacy localized—just fill in your own information. Place this repository in the skills directory of Claude Code (just call it interview-coach): After installation, click the "Three Steps to Get Started" below to fill in the information. All interview materials are stored by default in the directory you config.md specified. .gitignore ensures your real data won't be mistakenly sent back to the repository. 1. → Open the config.md and fill in the sessions_dir (where to store the session), user_name, and target_role. If you need to sync with Feishu, set lark_enabled to true and fill in the lark_space_id. 2. → Change the placeholders in the four files under Profile/ to your own real information (profile / Projects is required, talking-points is recommended, qa-bank does not need to be pre-filled). 3. → "Prepare for the interview at Company X, JD is ......", skill will start automatically. → Enter Mode A and generate 5 local files to /X Company / (if Feishu sync is enabled, documents will be created simultaneously) → Enter Mode C, generate three local files from 06-08 + automatically append to qa-bank.md (if you open Feishu, append to the end of the same company's document) - : In config.md, set lark_enabled to true, fill in the lark_space_id, and first complete lark-cli auth login - : The root directory of the Feishu knowledge base you configured - : One company, one document, document name = company abbreviation - : All rounds of research + review for the same company accumulated into the same article, using --- + timestamp segmentation - :p rofile/ personal information + 00-basic info.md (including salary/HR contact information and other sensitive information) - : For Mode A/C, say "Do not sync Feishu" / "skip Feishu"

Overview (machine translated)

General AI Job Interview Assistant — Suitable for any position and industry. All customized capabilities come from the information you fill in yourself; Skill itself contains no personal information.

This is a Claude Code Skill: before the interview, we help you conduct company research, JD analysis, Q&A prediction, and rhetorical question design. During the interview, you provide on-the-spot scripts, and after the interview, you do a recording review and weak point diagnosis. Ready to use out of the box, privacy localized—just fill in your own information.

Installation

Place this repository in the skills directory of Claude Code (just name the directory interview-coach):

After installation, click the “Three Steps to Get Started” button below to fill in your information. All interview materials are stored in the directory you specified by config.md by default, and .gitignore ensures your real information won’t be mistakenly sent back to the warehouse.


🚀 Start using it in three steps

  1. Fill in the configuration → Open config.md, then fill in sessions_dir (where to store sessions), user_name, target_role. If you need Feishu syncing, set lark_enabled to true and fill in lark_space_id.
  2. Fill in Information → Replace the placeholders in the four files under profile/ with your own real information (profile/projects required, talking-points recommended, QA-Bank not required to fill in).
  3. Speak plainly → “Prepare for the interview at Company X, JD is …,” and the skill automatically starts working.

⚠️ > Use it without filling out the data; skill can only be used as a blank template. The more detailed the filling, the more customized the output.

Directory Structure

How to trigger (speak plainly)

Scenario 1: Preparing for an interview

“Prepare for the interview at Company X. The position is, salary, JD content:…”

→ Enter Mode A and generate 5 local files to /X公司/ (if Feishu sync is enabled, the document will be created simultaneously)

Scenario 2: First aid during an interview

“During the interview, I was just asked, how should I answer?”

→ Enter Mode B and directly give ≤ 250-word script, no file generated

Scenario 3: Review after the interview

“Just finished talking with the HR at Company X, here’s the transcript:…”

→ Enter Mode C, generate three local files from 06-08 + automatically append to qa-bank.md (if you open Feishu, append to the end of the same company’s document)

Scenario 4: Viewing the pipeline

“Which ones are I facing right now? Which ones are I preparing next week?”

→ Enter Mode D and scan sessions/ to get the summary table


Feishu Sync Rules (optional, disabled by default)

  • How to start: Set lark_enabled to true in config.md, fill in lark_space_id, and complete lark-cli auth login first.
  • Location: The root directory of the Feishu knowledge base you configured

This page was translated automatically. Technical details may differ from the original.

Recommended Tools

Try a different keyword or remove a filter.

Install

npx skillfish add fenlili0108-source/interview-coach
New to MCP? Start with the mcpskill.org guide →