interview-practice
/interview-practice
Interactive mock-interview practice for a Full Stack .NET Developer role. Claude asks the candidate interview questions one at a time, waits for the answer, then scores it against a rubric and gives specific feedback plus a stronger model answer grounded in their real experience. Questions come from a researched bank covering C#/.NET, async, ASP.NET Core, Web API, security, EF Core, SQL Server, Angular/React, testing, architecture, Azure/DevOps, debugging, behavioural (STAR) and system design. Use whenever the user says "interview practice", "practice interview", "mock interview", "quiz me", "
- Author
- andreak3779 ↗
- Repository
- job-o-matic-assistant ↗
- Licence
- MIT
- Last repository change
- 12.11.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 12.11.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Interactive mock-interview practice for a Full Stack .NET Developer role. Claude asks the candidate interview questions one at a time, waits for the answer, then scores it against a rubric and gives specific feedback plus a stronger model answer grounded in their real experience. Questions come from a researched bank covering C#/.NET, async, ASP.NET Core, Web API, security, EF Core, SQL Server, Angular/React, testing, architecture, Azure/DevOps, debugging, behavioural (STAR) and system design. Use whenever the user says "interview practice", "practice interview", "mock interview", "quiz me", "
Triggers from the description
Original author description
Interactive mock-interview practice for a Full Stack .NET Developer role. Claude asks the candidate interview questions one at a time, waits for the answer, then scores it against a rubric and gives specific feedback plus a stronger model answer grounded in their real experience. Questions come from a researched bank covering C#/.NET, async, ASP.NET Core, Web API, security, EF Core, SQL Server, Angular/React, testing, architecture, Azure/DevOps, debugging, behavioural (STAR) and system design. Use whenever the user says "interview practice", "practice interview", "mock interview", "quiz me", "
How it works
- Step 1 — Set Up the Session
- Step 2 — Build the Question Set
- Step 3 — Load the Honesty Context
- Step 4 — Run the Practice Loop
- Step 5 — Wrap Up
- What This Skill Does NOT Do
SKILL.md preview
---
name: interview-practice
description: >
Interactive mock-interview practice for a Full Stack .NET Developer role. Claude
asks the candidate interview questions one at a time, waits for the answer, then
scores it against a rubric and gives specific feedback plus a stronger model
answer grounded in their real experience. Questions come from a researched bank
covering C#/.NET, async, ASP.NET Core, Web API, security, EF Core, SQL Server,
Angular/React, testing, architecture, Azure/DevOps, debugging, behavioural
(STAR) and system design. Use whenever the user says "interview practice",
"practice interview", "mock interview", "quiz me", "ask me interview
questions", "test me on .NET", "drill me", "help me practice answering", or
"practice behavioural questions" and has NOT given a specific job description.
If a job description is supplied, use interview-practice-job-description
instead. Do NOT use to generate the interview prep DOCX guide — that is
interview-prep. Output: live Q&A in the conversation, a session scorecard, and
an optional progress log.
---
# Interview Practice — Full Stack .NET Developer
A live mock interview. Claude plays the interviewer: asks one question, waits,
scores the answer, coaches, and moves on. Questions come from
`reference/interview-question-bank.md` (researched Oct 2026); scoring and the
session mechanics come from `reference/interview-practice-rubric.md`.
This is rehearsal, not document generation. Nothing here writes a DOCX, and
nothing writes to the vault's `profile/star-bank.md`.
> **Source files — load as indicated** (`reference/` is the repo-root `reference/` folder; `profile/` is the vault's profile folder):
> - `reference/interview-question-bank.md` — the question pool and the assessor's "Look for" key. **Never show a question's "Look for" text before the user has answered.**
> - `reference/interview-practice-rubric.md` — read in full before asking the first question; it defines the loop, scoring, feedback format and wrap-up.
> - `python -m jobomatic profile show facts` — the candidate's evidence tiers, differentiators and known gaps. Needed for the honesty check and for grounding model answers. Don't run `full` unless a behavioural model answer needs work-history detail.
> - the vault's `profile/gap-coaching.md` — load when a question touches a known gap or a 🟡/🔴/🕰️ skill (rubric §5).
> - the vault's `profile/star-bank.md` — load for `BE-` questions; model behavioural answers use these stories only.
> - `outputs/interview-practice/sessions.md` — optional; if it exists, read it to prioritise weak and untouched topics.
>
> **If any required file is missing:** stop and tell the user exactly which one, rather than improvising questions from memory.
>
> **Grounding rule:** every employer, project, course, number or outcome in a model answer must trace to the profile output or `star-bank.md`. If the evidence isn't there, give a generic best-practice answer *labelled as generic*. Never invent experience to make an answer sound stronger.
---
## Step 1 — Set Up the Session
Ask only for what you can't default. Offer these in one short message and accept
a one-line reply (or "just start"):
| Setting | Options | Default |
|---|---|---|
| **Focus** | Mixed · a topic (C#, async, ASP.NET Core, API, security, EF Core, SQL Server, front-end, testing, architecture, Azure/DevOps, debugging, behavioural, system design) · Lightning round | Mixed |
| **Level** | Junior `[J]` · Mid `[M]` · Senior `[S]` | Senior (the roles being targeted are senior; include `[M]` questions as the warm-up) |
| **Length** | Quick (5) · Standard (10) · Long (15+) | Standard (10) |
| **Mode** | Interview mode (no hints, one follow-up allowed) · Coach mode (hints and a short concept recap offered after each answer) | Interview mode |
If `outputs/interview-practice/sessions.md` exists, read it first and say in one
line how it will influence the mix (rubric §7).
If the user names a specific job posting or pastes a JD, stop and switch to
`interview-practice-job-description`.
## Step 2 — Build the Question Set
1. Read `reference/interview-question-bank.md`.
2. Select questions matching the focus and level; include entries one level below as warm-ups. For "Mixed", use roughly this spread for a Standard (10) session:
| Block | Count | Sections |
|---|---|---|
| Warm-up | 1–2 | `BE-01`/`BE-16` or a `[J]` question |
| Core .NET back end | 3 | §1–§4, §6 |
| Data, security or front end | 2 | §5, §7, §8 |
| Architecture / cloud / testing / debugging / AI | 2 | §9–§12, §16 |
| Behavioural (STAR) | 2 | §13 |
| System design | 0–1 | §14 (Senior, Long sessions) |
For a "behavioural" focus, vary the `BE-` pool across §13 (use the role-type table there); for a "questions to ask" focus, run `BE-17` and the §17 closing moves.
3. Prefer weak or untouched IDs from the session log. Never repeat an ID within a session.
4. Order from easier to harder; end on a behavioural or reflective question — ideally `BE-17` (their questions for the interviewer, scored per rubric §3 and the §17 pool).
5. **Do not print the list.** Tell the user the *shape* of the session ("10 questions: 2 warm-up, 5 technical, 2 behavioural, 1 design") and begin.
## Step 3 — Load the Honesty Context
Run the `facts` extraction once and keep it in mind for the whole session:
evidence tiers, Key Differentiators, Known Genuine Gaps, Resolved Framing Gaps,
Employer & Industry Context, Portfolio Projects. For any question you pick that
touches a known gap, also read the matching `gap-coaching.md` entry now so you
can assess calibration against its **Honest floor**.
## Step 4 — Run the Practice Loop
Follow `reference/interview-practice-rubric.md` §1 (ground rules) and §6
(feedback format) exactly. Per question:
1. **Ask** — state `Q{n}/{total}` and the question. For `SD-` prompts, give the scenario and ask the user to talk through requirements → design → trade-offs. Stop.
2. **Wait** for the answer. Handle `hint`, `skip`, `repeat`, `retry`, `stop` per the rubric.
3. **Probe once** if the answer is shallow or ambiguous. Skip the probe for `LR-` lightning questions.
4. **Score** with the technical rubric (§2) — or the STAR rubric (§3) for `BE-` questions. Apply the band (§4).
5. **Overclaim check** (rubric §5.3): compare any experience claim against the profile. If it's stronger than the evidence, say so kindly and offer the honest-floor phrasing.
6. **Give feedback** in the rubric §6 format: ✅ worked, ⚠️ missing/wrong, 🎯 stronger answer, 🔁 retry offer when < 3.5.
7. **Next question.** Don't ask whether to continue after every question — just continue, unless the user said `stop`/`pause`.
Coach mode only: after the feedback, add a 2–3 line concept recap and one link or search term for further reading.
For `BE-` questions the 🎯 model answer is built from a `star-bank.md` story
selected by its JD tags and "Use for" list; name the story, adapt the framing,
never change the facts. If no story fits, say that and coach the structure only.
## Step 5 — Wrap Up
When the questions are done (or the user stops), deliver the wrap-up in rubric
§7: scorecard table, topic averages, top 3 strengths, top 3 fixes, study list
(IDs < 3.5 with their bank sections), any honesty flags, and a next step.
Then offer, in one line, to append the session to `outputs/interview-practice/sessions.md`.
Only write it after a yes (or if the user earlier asked for tracking). The log
holds scores and IDs only.
Offer follow-ups where they fit:
- Weak topics linked to a real gap → the user may want `learning-plan-gap-analysis` or `primary-learning-plan-generator`.
- Preparing for a specific posting → `interview-practice-job-description`, or the `interview-prep` DOCX guide.
---
## What This Skill Does NOT Do
- Does not tailor questions to a job posting — use `interview-practice-job-description`.
- Does not generate the interview prep DOCX — use `interview-prep`.
- Does not write to the vault's `profile/star-bank.md` or any file other than the optional session log.
- Does not reveal answer keys before the user answers.
- Does not invent experience, projects or numbers for the user, and does not state cert status or course counts from memory — those come from the profile.
- Does not ask about or record protected-characteristic information.
- Does not replace live human practice; scores are a rehearsal aid, not a prediction.
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · andreak3779/job-o-matic-assistant | 85 | MIT |
| lingegrace/claude-research-skills | 100 | MIT |
Database snapshot: 13.11.2026 · package version 31a370cfd5b0 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 82interview-practice-job-description
Job-specific mock-interview practice. Takes a job description, works out what that employer will probably ask, and runs a live Q&A: one question at a time, scored against a rubric, with feedback and model answers grounded in the candidate's real experience. The question set is… - 100jobhunt-interview
Runs a realistic mock interview for one job with a panel of interviewer agents (recruiter screen, hiring manager, bar raiser): behavioral STAR questions plus role-specific ones built from the posting. Asks one question at a time, scores each answer against a five-part rubric,… - 90interview-prep
Builds a role-specific interview prep kit for any field - likely questions predicted from the JD (behavioral, technical/functional, situational, role-play, case), STAR stories mapped from the candidate's real experience, answers for gaps and tricky questions, and questions to… - 87mock-interview
Run a realistic mock job interview for developers and vibe coders, grounded in the project in the current folder (or a chosen role and stack when there is no project). Asks one question at a time across technical, project deep-dive, code & algorithms (explain key lines, trace an… - 90mock-interview
Run a realistic mock interview for a product marketing role. Plays a specific interviewer (recruiter, hiring manager, product partner, exec), asks one question at a time, pushes with follow-ups, then scores each answer and gives direct feedback with a stronger version. Use for… - 86interview-coach
When user asks for interview prep, mock interview, practice questions, behavioral questions, technical interview, HR round, salary negotiation, STAR method, common interview questions, company research, interview tips, confidence building, answer feedback, body language tips,…