How do I install the Interview Persona skill?

Run npx skillmds@latest add alaimo-labs/interview-persona in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Interview Persona safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Interview Persona?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Interview Persona free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Interview Persona?

alaimo-labs (@alaimo-labs) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Interview Persona by alaimo-labs — Agent Skill

Name: Interview Persona
Rating: 4.2 (5 reviews)
Author: alaimo-labs

File contents

/interview-persona

Run a synthetic user interview, following the synthetic-interviews skill.

Input: $ARGUMENTS

Workflow

Resolve the persona. Find the named persona in product/personas/. If no name given, list available personas (with their type) and ask. If none exist, suggest /generate-personas first. If the persona has no type: line (created before the field existed), propose one from its content and offer to write it into the file before starting — with the interviewer's approval; without it, proceed and treat the type as unknown.
Resolve the mode. Exploration (open discovery) or validation (feedback on a specific idea). If not specified and not obvious, ask — one question, two options. In validation mode, also ask what idea/prototype/direction is being validated if not provided.
Load product context from product/overview.md (or equivalent).
Set the scene in one short message: who the persona is (one line, type included), the mode, and that the user is now the interviewer. Then hand over — they ask the first question.

How this evidence reads, by type. Interviewing a negative persona is useful — they say why not, and that bounds the scope — but their answers count neither for nor against a value belief of the target segment. A tertiary persona informs viability (who pays, approves, blocks), not desirability: their answers do not confirm or contradict value beliefs either. Say this in one line when the persona is negative or tertiary, and repeat it in the transcript header so /extract-insights reads it too.
Roleplay per the synthetic-interviews skill and the mode-specific instructions (exploration-mode.md / validation-mode.md). Stay in character until the interviewer ends the interview. Interviewer coaching notes go outside the roleplay, clearly marked, and sparingly.
On wrap-up: offer to save the transcript to product/interviews/{YYYY-MM-DD-HHMM}-{persona-slug}.md (header: persona, type: next to source: synthetic, mode, topic, date — plus the one-line reading warning when the persona is negative or tertiary) and suggest /extract-insights on it.

Language

Conversation and the transcript in the language of the conversation. The type: and source: values are fixed English tokens in every language.

---name: interview-personadescription: Interview one of your synthetic personas — exploration mode (open discovery) or validation mode (feedback on an idea)---# /interview-personaRun a synthetic user interview, following the `synthetic-interviews` skill.Input: $ARGUMENTS## Workflow1. **Resolve the persona.** Find the named persona in `product/personas/`. If no name given, list available personas (with their type) and ask. If none exist, suggest `/generate-personas` first. If the persona has no `type:` line (created before the field existed), propose one from its content and offer to write it into the file before starting — with the interviewer's approval; without it, proceed and treat the type as unknown.2. **Resolve the mode.** Exploration (open discovery) or validation (feedback on a specific idea). If not specified and not obvious, ask — one question, two options. In validation mode, also ask what idea/prototype/direction is being validated if not provided.3. **Load product context** from `product/overview.md` (or equivalent).4. **Set the scene** in one short message: who the persona is (one line, type included), the mode, and that the user is now the interviewer. Then hand over — they ask the first question. **How this evidence reads, by type.** Interviewing a `negative` persona is useful — they say why not, and that bounds the scope — but their answers count neither for nor against a value belief of the target segment. A `tertiary` persona informs viability (who pays, approves, blocks), not desirability: their answers do not confirm or contradict value beliefs either. Say this in one line when the persona is `negative` or `tertiary`, and repeat it in the transcript header so `/extract-insights` reads it too.5. **Roleplay** per the `synthetic-interviews` skill and the mode-specific instructions (exploration-mode.md / validation-mode.md). Stay in character until the interviewer ends the interview. Interviewer coaching notes go outside the roleplay, clearly marked, and sparingly.6. **On wrap-up:** offer to save the transcript to `product/interviews/{YYYY-MM-DD-HHMM}-{persona-slug}.md` (header: persona, `type:` next to `source: synthetic`, mode, topic, date — plus the one-line reading warning when the persona is `negative` or `tertiary`) and suggest `/extract-insights` on it.## LanguageConversation and the transcript in the language of the conversation. The `type:` and `source:` values are fixed English tokens in every language.

Frequently asked questions

Interview Persona