- Home
- Skills
- Prompt & Skill Engineering
- Discovery Interview Brief — Structured Product-Agent Intake
Discovery Interview Brief — Structured Product-Agent Intake
Free AgentOps Lab skill: turn a messy research signal into an interview brief with a learning goal, who to talk to, and non-leading questions. Not a PMF or income guarantee.
Free
Works with the AI Agents you already use
Discovery Interview Brief — Structured Product-Agent Intake
Example session with this skill installed
Create a one-screen Discovery Interview Brief from this authorized, consented research signal. Use only the text in brackets; do not infer missing facts or add quotes: [PASTE A SANITIZED SIGNAL HERE]. No personal or confidential details. Identify interviewees only through owned or authorized channels. Separate facts from assumptions; ask no more than five non-leading questions. If consent or research authority is unclear, stop and ask for clarification.
- Read your context and instructions
- Compiled the discovery interview brief
Illustrative format — placeholders, not evidence
Restated signal: [One sentence faithful to the supplied signal; if none, say “Not provided.”]
One learning goal: [A single question the research should answer, derived only from the signal.]
Who to interview: [Relevant role or relationship supported by the signal.] Reach people only through owned or otherwise authorized channels, with consent. If unsupported, say “Not specified.”
In scope: [Topics directly needed to answer the learning goal.]
Do not ask: [Unnecessary sensitive or identifying details; anything outside the goal.]
Session shape: [A practical format and duration to confirm with the research owner; no invitation is sent.]
Consent and privacy check: Explain purpose, voluntary participation, note-taking or recording, retention, and how findings will be used. Obtain permission before collecting notes or recording. If authorization or consent is unclear, pause. Sanitize PII and secrets.
Non-leading questions (up to five):
- Can you walk me through the last time [situation from the signal] happened?
- What did you try next?
- What, if anything, made that difficult?
- What information did you need at that point?
- What else should I understand about that experience?
Next action: The owner supplies a sanitized signal, confirms consent and research authority, then reviews this brief before any interview is arranged. This template creates no participants, findings, quotes, or outcomes.
Connects securely to your tools. The creator never sees your data.
About this skill
You already use agents on discovery work. What still breaks: messy signals that never become a reusable interview brief, and questions that start before the learning goal or consent path is clear.
Discovery Interview Brief by AgentOps Lab turns a messy signal — a support thread, a user message, a sales note, an observation — into a one-screen brief: restated signal, one primary learning goal, who to interview, in-scope vs do-not-ask topics, session shape, ethics and consent checks, and clear non-leading questions.
Compliance: Structure interview briefs only. Consented people, or users of products you own / are authorized to research. Not a survey-spam or cold-outreach kit. Do not invent interview quotes, participants, or fake evidence. No scraping private data. No deanonymizing. Sanitize PII and secrets. No PMF, fundraising, revenue, or “you will ship a hit” guarantees. You remain responsible for research ethics, privacy law, and what you ship.
Install as markdown SKILL.md for Cursor / Claude Code / Grok Bot-style agents.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Prompt & Skill Engineering.
Limited to 5 questions per brief. Does not generate outreach lists or scrape data. Requires user to have authorization for the research subjects.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 5 days ago
- Free with an Agensi account
Requires
Nothing to install.
Permissions
No special permissions declared or detected