Stakeholder readout generator

A 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

Download the skill

Enter your details to download the skill file.

What it does

A stakeholder readout is the version of the research other people actually read: findings stated plainly, ordered by what the audience has to decide, with evidence attached. It gets called a research share-out, a debrief, a topline, or just the deck, depending on the team. The failure mode is always the same. Findings get flattened into assertions and the evidence stays in a folder nobody opens. Six weeks later somebody asks where a number came from and nobody can answer.

This skill builds the readout from a study that already exists in your Great Question repo. It reads the sessions, transcripts and highlights from that study, writes the findings, and keeps each one attached to the highlights it came from, so a stakeholder who doubts a finding can watch the moment it rests on. Traceability is the point of running this against the repo rather than a folder of documents. It works from Claude, Cursor or ChatGPT once the Great Question MCP is connected, and what it produces lands as an insight in your repo, the object that already holds your findings and stories.

What you get back

Findings written from the study's own sessions, ordered for the audience you name.
The session highlights behind each finding, linked so anyone can open the source.
The participants each finding rests on, and how many of them there were.
Open questions the study did not answer, kept in rather than quietly dropped.
A readout saved into your Great Question repo as an insight, sitting with the rest of your findings.

Install it in Claude, ChatGPT, Copilot or Gemini

The same skill file works in all four. Pick your tool for the steps.

1
Connect the Great Question MCP to Claude
These skills act inside your Great Question account, so the MCP has to be connected first. The MCP guide has the setup for each tool.
2
Download the skill
Fill in the short form to get the .skill file.
3
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.
4
Just ask
Ask in plain language. Claude picks up the skill on its own when the request matches.

How to use it

Once it's installed, ask in plain language. Your AI follows the skill's method when the request matches.

Example prompt 1

Build the readout for the reconciliation study, audience is the payments PM and her eng lead.

Example prompt 2

Same study, but a five-minute version for the exec review on Thursday.

Example prompt 3

Which findings only rest on one participant? Flag them in the readout.

What it won't do

It won't decide what your team should do about the findings, and it won't manufacture confidence in a thin one. A finding that rests on two people is reported as resting on two people. It also references highlights rather than creating them, so a study nobody has clipped yet gives it less to work from, and sessions that never made it into the repo are invisible to it.

Questions

How is this different from the stakeholder readout Claude skill?

The Claude skill works on files you hand it and needs no account, which suits work that lives outside the platform. This one reads the actual study in your repo and links every finding back to the session highlight it came from. Same output shape, different evidence trail.

Where does the readout end up?

As an insight in your Great Question repo. Insights are the object that holds research findings and stories in the platform, so the readout sits with the studies and sessions it was built from and stays searchable next to the rest of your work.

Does it use highlights my team made, or does it make its own?

It uses the ones your team already made. There is no operation for creating a highlight, so the clips people took while reviewing sessions are the evidence it reaches for. A study nobody has been through yet gives it less to work with.

More skills

See all skills