Claude skills for user research
We run research too, and we got tired of explaining the same method to our AI every time. So we wrote them down. Every skill here is tested by our team on our own studies, and each one installs in Claude, ChatGPT, Microsoft Copilot or Gemini, whether you're a UX researcher, a PM, in research ops or design.
Pick a skill below and fill in the short form. The download is ready straight away.
- 26
- skills published
- Weekly
- new drop
- Free
- no account needed
Research skills
These work on files you already have: transcripts, notes, survey exports. No account needed, in whichever AI tool you use.
Evidence integrity auditor
Works on its ownA Claude skill that checks a research report against its transcripts for quotes that can't be found, findings that rest on one person, and numbers that mislead.
AnalysisOctober 12, 2026
Instrument scorer
Works on its ownA Claude skill that scores standard usability questionnaires like SUS correctly, compares them to the benchmark, and checks your questions still match the standard wording.
AnalysisOctober 12, 2026
Moderator scorecard
Works on its ownA Claude skill that shows you, from your own interview transcript, how much you talked, interrupted or led the participant, so you can get better at moderating.
Research opsOctober 12, 2026
Panel data steward
Works on its ownA Claude skill that finds messy participant data, like one job title stored four different ways, and proposes fixes that a person approves before anything changes.
Research opsOctober 12, 2026
Sample size planner
Works on its ownA Claude skill that tells you how many participants your study needs, with the published source behind the number and a sentence you can paste into your plan.
Study designOctober 12, 2026
Screener porter
Works on its ownA Claude skill that checks a recruitment screener before launch and helps you copy it into another study or tool without losing which answers qualify people.
RecruitingOctober 12, 2026
Task scenario linter
Works on its ownA Claude skill that checks your usability test tasks for wording that gives the answer away, such as naming the button, before you run the test.
Study designOctober 12, 2026
Theme quality auditor
Works on its ownA Claude skill that checks whether your synthesis themes are real findings or just topic labels, and asks a question to help you turn each label into a finding.
SynthesisOctober 12, 2026
Affinity mapper
Works on its ownA Claude skill that clusters raw research observations into themes bottom-up, then tells you which of those themes you should not trust yet.
SynthesisAugust 20, 2026
Research synthesizer
Works on its ownA Claude skill that reads across several studies at once and reports where they contradict each other, keeping the study and the participant attached to every claim it makes.
SynthesisAugust 13, 2026
Discussion guide builder
Works on its ownA Claude skill that turns research objectives into a semi-structured guide, with every question traced back to the objective it serves and the leading ones rewritten before you see them.
Study designAugust 6, 2026
Insight tagger
Works on its ownA Claude skill that tags research highlights against your codebook and holds a written definition for every code, so you can see which quotes it could not confidently place.
AnalysisJuly 30, 2026
Stakeholder readout
Works on its ownA Claude skill that turns findings sitting in your files into a readout ordered by the decisions it should inform, with the evidence for each headline attached underneath.
SynthesisJuly 23, 2026
Screener builder
Works on its ownA Claude skill that turns a persona description into screener questions with qualifying logic, written so nobody can tell which answer gets them into the session.
RecruitingJuly 16, 2026
JTBD analyzer
Works on its ownA Claude skill that codes interview transcripts against the Jobs to be Done framework, returning the four forces and the moment someone actually switched, with the quote behind each one.
AnalysisJuly 9, 2026
JTBD interview guide builder
Works on its ownA Claude skill that writes a full JTBD moderation guide around the Moesta timeline interview, including the probes that get someone back to what actually happened.
Study designJuly 2, 2026
Research brief writer
Works on its ownA Claude skill that turns a business question into a research brief, naming the method to run and the decision waiting on the result, with the limits written in.
Study designJune 25, 2026
Survey designer
Works on its ownA Claude skill that turns a stated research goal into a survey you can actually analyse, with the reasoning for every question written next to the question.
Study designJune 18, 2026
Synthetic user
Works on its ownA Claude skill that builds reusable customer profiles from clusters of real interview evidence, with every attribute cited to the session it came from and every gap marked.
SynthesisJune 11, 2026
Transcript cleaner
Works on its ownA Claude skill that turns raw Otter, Fireflies, Granola and Zoom exports into transcripts you can analyse, with speakers resolved and timestamps kept where quoting needs them.
Research opsJune 4, 2026
Great Question skills
These do the work in your account, not just the draft. Describe a study and it gets built. Ask for a readout and every finding comes back with the session highlights behind it. Connect the MCP once and they run from Claude, ChatGPT, Copilot, Gemini or Cursor.
Interview study builder
Needs the Great Question MCPA Great Question skill that takes a description of the study you want and builds it in your account, where it waits for your review before it goes anywhere.
Study designAugust 27, 2026
Panel hygiene auditor
Needs the Great Question MCPA Great Question skill that audits your panel for duplicate records and contacts that have gone quiet, working from consent dates, unsubscribes and participation history against a threshold you name.
Research opsAugust 27, 2026
Recruiter
Needs the Great Question MCPA Great Question skill that screens your own panel against the criteria on a study you have already built, then gets the people who qualify invited.
RecruitingAugust 27, 2026
Stakeholder readout generator
Needs the Great Question MCPA Great Question skill that builds the readout from a finished study in your repo, with the real session highlights behind every finding so anyone can check the evidence themselves.
SynthesisAugust 27, 2026
Study email writer
Needs the Great Question MCPA Great Question skill that writes the messages a study needs and drafts them against that study, so participant comms stop being the thing nobody wants to do at 6pm.
RecruitingAugust 27, 2026
Unmoderated test builder
Needs the Great Question MCPA Great Question skill that turns a concept or a task you describe into a working unmoderated study in your account, built and waiting for you to check it.
Study designAugust 27, 2026
AI research workflows
The jobs research teams actually run against their own data through Claude, Cursor and ChatGPT. Each one names the tools it calls and where you sign off, so you can see what the agent touches before you let it near anything.
- 01
The daily interview digest
Finds every session that completed today across your active studies and posts a short summary where the team already looks.
Read onlySchedule daily - 02
The transcript vault
Copies completed transcripts into your team's document store, one document per session, named so they are findable six months later.
Writes outside GQSchedule weekly - 03
Transcript to markdown
Turns a session transcript into clean markdown you can drop into a repo, a wiki or your notes, without paraphrasing a word.
Read only - 04
Attach the evidence to the ticket
Pulls the transcript and the participant's company onto the matching task in your planning tool, so whoever picks up the work can read the source.
Writes outside GQ - 05
The response rate tracker
Counts how many people you invited against how many submitted a screener, for any date range, without building a report.
Read onlySchedule monthly - 06
Screener cloning
Lifts screener questions you already trust from a past study into a new one, then reads the result back before anything goes out.
You approve first - 07
Cross-study quote mining
Builds a corpus across the studies you name, sorts participants by segment, and pulls verbatim quotes with the session they came from.
Read only - 08
The highlight reel shortlist
Finds the clips worth showing, with exact timestamps, speaker attribution and a link straight to the moment.
Read only - 09
Panel feasibility check
Tells you how many people in your panel match the criteria, before you commit to a study you cannot recruit for.
Read only - 10
CRM to panel matching
Checks in bulk whether the people in your CRM are already research participants, so you stop inviting the same person through two doors.
Read only - 11
Bulk panel tagging
Applies the same attribute across your panel so one filter returns everyone it should.
You approve first - 12
Study email localisation
Translates your study email templates for a second region without rewriting each one by hand, and replaces nothing until you approve.
You approve first
GQ Skills · Governed AI research Or write your own
Everyone can run research now. Someone has to keep it good. GQ Skills are your team's standards, made enforceable: screener quality, spend caps, tone. Write a skill once and it fires on its own, on the step it governs.
Because enforcement lives on our server rather than in the AI's context, it holds wherever the work starts: our app, Claude, Cursor, Copilot, Gemini. A skill sitting in a repo can be forgotten or skipped. This one can't.
Standard
Proven research craft, ready to turn on. Start here if you don't have a house method written down yet.
Custom
Your way of running research, written down once. Editable and granular, so you can scope a skill to the exact step it governs.
Required
Can't be skipped. The tool call doesn't complete until the skill has run, so a screener that breaks your rules never ships.
- Verify any claim in one click
- No real names exposed, ever
- Says “not found” instead of guessing
A prompt is a suggestion. A skill is a control.
How to use a skill
A skill is a set of instructions your AI reads when the task matches. Install it once in the tool you use, then ask in plain language and it follows the method. Pick your tool for the steps.
Install in Claude
- 1
Download the skill
Fill in the short form to get the .skill file.
- 2
Add it to Claude
In the Claude app go to Settings → Capabilities → Skills and upload the file. In Claude Code, run the npx command shown after the form, or drop the folder into your project's skills folder.
- 3
Just ask
Ask in plain language. Claude picks up the skill on its own when the request matches.
Install in ChatGPT
- 1
Download and unzip the skill
The .skill file is a zip. Inside is SKILL.md, the instructions, plus any reference files the skill uses.
- 2
Make a Custom GPT or a Project
Create a Custom GPT (Explore GPTs → Create → Configure) or a Project. Paste SKILL.md into its instructions and upload the reference files as knowledge.
- 3
Just ask
Open that GPT or Project and ask in plain language. It follows the skill's method every time.
If the instructions box won't take the whole of SKILL.md, upload SKILL.md as a knowledge file instead and write one line of instructions: follow SKILL.md for every request.
Install in Microsoft Copilot
- 1
Download and unzip the skill
The .skill file is a zip. Inside is SKILL.md, the instructions, plus any reference files the skill uses.
- 2
Build a Copilot agent
In Microsoft 365 Copilot, open Agent Builder and create an agent. Paste SKILL.md into its instructions and add the reference files as knowledge.
- 3
Just ask
Open the agent and ask in plain language. Share it with your team so everyone runs the same method.
If the instructions box won't take the whole of SKILL.md, upload SKILL.md as a knowledge file instead and write one line of instructions: follow SKILL.md for every request.
Install in Gemini
- 1
Download and unzip the skill
The .skill file is a zip. Inside is SKILL.md, the instructions, plus any reference files the skill uses.
- 2
Make a Gem
In Gemini, go to Gems → New Gem. Paste SKILL.md into its instructions and add the reference files as knowledge.
- 3
Just ask
Open the Gem and ask in plain language. It follows the skill's method every time.
If the instructions box won't take the whole of SKILL.md, upload SKILL.md as a knowledge file instead and write one line of instructions: follow SKILL.md for every request.