ai-interviewer — Claude Code subagent: interview and setup
The ai-interviewer Claude Code subagent runs a practice interview against a revised resume and scores answers against job criteria. It probes specific resume claims, then reviews the transcript.
no license ★ 73 updated 2026-11-11
/plugin marketplace add Vets-Who-Code/ai-era-job-kit
/plugin install ai-era-job-kit@ai-era-job-kitWhat ai-interviewer does
The subagent creates a rubric from the job posting before asking questions, then conducts the interview one question at a time. Each score needs supporting words from the transcript, and weighted criteria feed into the final result. The rubric is saved before the first question and stays fixed once the interview begins.
It flags claims that go beyond the resume, lines the candidate could not defend, and answers that sound rehearsed. It also checks for missing parts of STAR-L and asks candidates to replace vague “we” statements with a clear account of their own actions. Accent, tone, fluency, pace, pauses, personality, and confidence do not affect scores.
What it helps with
- test whether a candidate can defend each resume line
- compare interview answers with job criteria
- spot claims that go beyond the resume
- find missing parts of a STAR-L answer
Example prompts for the agent
- Run a practice interview using my revised resume and this job posting.
- Ask one question at a time and challenge specific claims on my resume.
- Score this interview transcript against the job criteria and cite evidence for each score.
How to install ai-interviewer
1Plugin marketplace
/plugin marketplace add Vets-Who-Code/ai-era-job-kit/plugin install ai-era-job-kit@ai-era-job-kit2Manually: put the files in place
.claude/agents/ai-interviewer.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/ai-interviewer.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/ai-interviewer.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about ai-interviewer
How do I install ai-interviewer in Claude Code?
Run /plugin marketplace add Vets-Who-Code/ai-era-job-kit, then /plugin install ai-era-job-kit@ai-era-job-kit. The agent file is .claude/agents/ai-interviewer.md.
What does ai-interviewer do?
It conducts a practice interview based on a resume and job posting, then scores the transcript against a rubric with quoted evidence. It also flags unsupported claims and weak answers.
Does ai-interviewer score accent or tone?
No. It leaves accent, tone, fluency, pace, pauses, personality, and confidence out of the scores.