track-questions
Skill0 starsChanged 4 days agoMITView on GitHub
Manage the buyer questions supereo tracks for a brand on ChatGPT, Gemini, Google AI Mode, AI Overviews, Perplexity and Claude. Lists the tracked questions with their demand (the Google searches closest to each, about how many a month, the trend and a band), estimates demand for new ones, adds, rewrites or removes questions, and asks one again now. Use when someone wants to track a new buyer question, asks which questions supereo tracks or how often buyers search for them, or wants to change their tracked panel.
In this repository4 other files it gives its agents.
- aeo-auditSkill
- ai-visibility-auditSkill
- ai-visibility-resultsSkill
- ship-recommendationSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 21
When the person has several sites (list_workspaces), pass the site as site to every tool below.
- Distinctive: line 27
Every change makes a new panel version.
- Distinctive: line 36
Write each one the way a buyer asks an AI: a whole question from the buyer's side ("Which email API should a small SaaS team use for …
- Distinctive: line 37
Leave the brand's name out of a question meant to measure how often AI recommends it.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Track buyer questions
Tools: list_workspaces, list_questions, list_answers, estimate_demand, add_question, add_questions, replace_question, remove_question, rerun_question and get_next_step, from supereo's MCP server. estimate_demand and rerun_question are paid; the rest are free.
supereo measures how often AI recommends a brand on a panel of buyer questions. This skill reads that panel and changes it, the way the Questions page in supereo does.
1. Get the tools
They come from supereo's MCP server, https://supereo.app/api/mcp, signed in as the person. Some clients load MCP tools only when asked: search your tools for list_questions before you decide it isn't there. If it isn't, connect the server:
- Claude Code:
claude mcp add --scope user --transport http supereo https://supereo.app/api/mcp, then/mcpto sign in. The tools load in the next session. - Claude (claude.ai, the desktop app, and Claude Code signed in with a Claude account): add it as a connector with this link, then connect and sign in.
- Any other agent: https://supereo.app/install has the steps for each one.
When the person has several sites (list_workspaces), pass the site as site to every tool below.
2. How the panel works
- A site supereo tracks has a panel of up to 40 buyer questions, each asked on all six engines three times a week. Its rates come from the panel's last 4 weeks of answers.
- Tracking starts by itself once the brand has a supereo plan. A site with only its free audit has the audit's questions, which
list_questionslists without ids, and no panel to change:get_next_stepgives the person the link to start a plan. - Every change makes a new panel version. Only the questions it touches start again; every other question's counts carry on.
- A question that doesn't name the brand counts toward how often AI recommends it. One that names the brand reports apart, with the branded questions.
3. Read the panel
list_questions: each question with its id, its position, whether it names the brand, its last 4 weeks of answers, and its demand: the Google searches closest to it, about how many searches they get a month together, their 12-month trend, and a band (very low under 50 a month, low 50 to 500, medium 500 to 5,000, high over 5,000). list_answers with a question_id gives that question's answers in full.
4. Choose new questions
- Write each one the way a buyer asks an AI: a whole question from the buyer's side ("Which email API should a small SaaS team use for password resets?"), never a keyword.
- Leave the brand's name out of a question meant to measure how often AI recommends it.
- Before you add them, read their demand with
estimate_demand, up to 20 candidates at once: each one's closest searches, their monthly total, the trend and the band. It's paid: it runs with a plan and counts against the account's daily allowance, so say so and get the person's yes first. - Say what demand is: a directional read of Google searches, never a count of questions asked of AI. Put it as "about 1,300 Google searches a month for the closest searches", name those searches, and say when they were read.
5. Change the panel, with a yes first
Show the person exactly what will change and get their yes before each change. Then:
add_question: one question. supereo asks it on all six engines now, then three times a week like the rest.add_questions: several at once, as one version. Nothing is added when one of them can't be (a repeat, one already on the panel, or past 40): the answer names which.replace_questionwith itsquestion_id: rewrites one. The old question retires, its answers still readable by its id, and the new text starts a new series in its place.remove_questionwith itsquestion_id: supereo stops asking it, and rates stop counting it. A test that holds it keeps asking it until the test ends.
Each one answers with the demand of the questions it adds: pass that on to the person.
6. Ask one again now
It's paid, so say which question and which engines it would ask, and get the person's yes first. Then rerun_question with its question_id, and engines to name some, or all six: the new answers show on the question as reruns, and never count in a rate. An engine that asked it in the last 30 minutes is skipped, and the answer says when it can run again. It takes a minute or two, and counts against the account's daily allowance.
Where to go from here
- What the answers say and who the engines recommend: the ai-visibility-results skill.
- Make one of supereo's recommended changes yourself: the ship-recommendation skill.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

