Tailored interview questions
Skill ur-grue/autopunk-media-skills/skills/pre-production/tailored-interview-questions
Generates a set of interview questions precisely matched to a named subject's known background, public record, and the specific story angle — so every question earns its place and no time is wasted on answers the journalist already has.From its SKILL.md
npx -y skills add ur-grue/autopunk-media-skills --skill tailored-interview-questionsAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 21 stars21 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
8.0 KB, ~1.5k tokens by cl100k_base, as published. Nobody here has run it
Tailored Interview Questions
What This Skill Does
Generates a set of interview questions precisely matched to a named subject's known background, public record, and the specific story angle — so every question earns its place and no time is wasted on answers the journalist already has.
When To Use This Skill
- You have confirmed an interview and have done initial research but need to turn that research into sharp, ordered questions
- Your subject has a detailed public record (previous interviews, published work, official statements) and you want to probe beyond what they usually say
- You are preparing for a high-stakes or time-limited interview and cannot afford vague or redundant questions
- You need to ensure questions serve the story angle rather than simply covering the subject's biography
What You Need To Provide
Required: The subject's name and role, a brief summary of their relevant background and public record, and the story angle or editorial purpose of the interview. Optional: Format (on-camera, radio, print), available interview time, any topics that are off-limits, previous interviews or public statements you want to challenge or follow up on, specific claims you need the subject to respond to.
How the Assistant Approaches This
- Reviews the subject's provided background to identify what is already known, what is on the public record, and where gaps or contradictions exist — these gaps become the most valuable questions.
- Maps the questions to the story angle, discarding anything that is interesting about the subject in general but irrelevant to this specific story.
- Orders the questions to build rapport early, move into substantive territory in the middle section, and place the hardest or most confrontational questions toward the end — after trust is established and the recorder is running.
- Flags any question that requires the subject to contradict a prior statement, so the journalist can have the original source to hand during the interview.
Output Format
A numbered list of 10–20 questions, grouped into three sections: opening (2–3 questions to establish context and tone), core (the bulk, directly serving the story angle), and closing (1–2 questions inviting reflection or a final statement). Each question is followed by a one-line note explaining its purpose or what a good answer would reveal. Plain language throughout — questions should be speakable, not written.
Quality Criteria
- Every question is specific to this subject — none could be asked of any expert in the field without modification
- At least one question directly addresses a gap, contradiction, or evasion in the subject's public record
- Questions are ordered so the interview builds in intensity without alienating the subject early
- No question contains two sub-questions — each is singular and answerable
- The story angle is visible throughout — a reader could identify the editorial purpose from the questions alone
- Questions are conversational in register, not formal or legalistic
Example
Input
Subject: Dr. Marina Okonkwo, Chief Medical Officer of a regional public hospital network. Background: She oversaw a restructuring of the network's emergency department two years ago, which was presented publicly as an efficiency improvement. Since then, average A&E waiting times have increased by 34% according to the network's own published data. She gave a conference presentation eight months ago in which she said the restructuring had "exceeded its targets." She has not previously spoken to the press about the waiting time figures. Story angle: Whether the restructuring prioritised cost savings over patient outcomes, and why the public data contradicts the CMO's public statements.
Output
Interview Questions: Dr. Marina Okonkwo Story angle: A&E restructuring — cost savings vs. patient outcomes
Opening
-
Can you describe what the emergency department looked like before the restructuring and what specific problems you were trying to solve? Purpose: Establishes her framing and gives her room to speak freely before harder questions arrive.
-
What were the original targets the restructuring was designed to meet, and who set them? Purpose: Gets her targets on the record so they can be measured against outcomes.
-
Who else was involved in the design of the restructuring — clinical staff, administrators, external consultants? Purpose: Establishes accountability and whether frontline staff had input.
Core
-
The network's published performance data shows average A&E waiting times have increased by 34% in the two years since the restructuring. How do you account for that? Purpose: The central question — her response will reveal whether she accepts the data or contests it.
-
At the Regional Health Leadership Conference last year, you said the restructuring had "exceeded its targets." Which targets were you referring to specifically? Purpose: Directly tests the apparent contradiction between her public statement and the published data.
-
Was reducing waiting times one of the restructuring's stated targets? Purpose: If it was not, that is a significant editorial finding. If it was, her conference claim requires explanation.
-
Were nursing or administrative staffing levels reduced as part of the restructuring? Purpose: Probes the cost-saving dimension without leading.
-
Have emergency department staff raised concerns with you about the impact on patients since the restructuring? If so, through what channels? Purpose: Assesses whether internal warning signs existed and whether they reached her level.
-
Is a 34% increase in waiting times, in your clinical judgement, an acceptable outcome? Purpose: Forces a value judgement on the record.
-
Has the board reviewed the waiting time figures? What was their response? Purpose: Extends accountability upward.
-
Has the restructuring model been reviewed or revised in light of the performance data? Purpose: Establishes whether the organisation is responsive to evidence.
-
If you were designing the restructuring today with the benefit of hindsight, what would you do differently? Purpose: Opens a path for accountability without requiring a direct admission of failure.
Closing
-
Is there anything about the restructuring and its outcomes that you believe the public should understand that has not been reported? Purpose: Gives her the final word and sometimes produces the most candid answer of the interview.
-
What would a successful emergency department look like at this network in five years? Purpose: A forward-looking close that ends on constructive ground and may produce a quotable commitment.
Known Limitations
- The quality of the questions depends entirely on the quality of the background information provided. Thin research produces generic questions — this skill cannot substitute for pre-interview reading.
- The assistant has no access to the subject's actual previous interviews or statements. You must supply the relevant quotes and claims; the assistant will then build questions around them.
- For subjects who are likely to be legally cautious or media-trained to deflect, the assistant can note where questions are likely to be evaded, but cannot predict or counter specific deflection strategies.
Related Skills
What ships with it: 1 file
94 B alongside SKILL.md