
QuizGen MCP server
Which MCP servers are QuizGen's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.
read 2026-10-28
- Registry namedev.quizgen/quizgenverified in the MCP registry
The registry name sits under quizgen.dev. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-10-28
- Tools14
- Server calls itselfquizgen
- Server version1.0.0
- Protocol version2025-06-18
On 2026-10-28 the server answered initialize and tools/list without a sign-in. First answer in 185 ms.
- Its tools do every use case we checked: collect client intake answers before a first call; build an estimate form for a house painting job.
Once connected
- For "collect client intake answers before a first call", the likely first call is create_quiz.
- For "build an estimate form for a house painting job", the likely first call is create_quiz.
- Keep approval on for send_quiz_by_email: they delete, send or pay without saying so.
- create_quiz, update_quiz, delete_quiz, set_webhook, send_quiz_by_email change data without a sign-in: anyone with the address can call them.
- Parameters are untyped or undescribed in 5 of 14 tools: check arguments before calling.
Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.
- INITE MCP score64 of 100answers 100 · does the job 100 · definitions 94 · context 63 · connects 100 · lowered by the warning signs below
- What an agent can do with it2 of 2 use cases
- Answered100% of days7 of 7 daily checks
- Median first answer167 ms
- Space in an agent's context≈ 11,801 tokensthe definitions of all its tools (14), which an agent carries on every turn
- Tool definitionsA · 94 of 100
- Typed results14 of 14 toolsdeclared outputSchema: results come typed, read once
- Parameters without a description: in 5 of 14
What an agent can do with it
- can do · collect client intake answers before a first call
- not sure · score leads so good prospects rise to the top
- can do · build an estimate form for a house painting job
Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?
Worth a look
- Tools that delete, send or pay, marked as not destructive: send_quiz_by_email
- Tools that change data, on a server that answers without a sign-in: create_quiz, update_quiz, delete_quiz, set_webhook, send_quiz_by_email
A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →
create_quizCreate a live, hosted quiz/form/survey and get its shareable URL. Without authorization this creates a temporary sandbox quiz (72 hours, 10 responses) with a secret claim_url to keep it. The definition schema below is the complete quiz…
list_quizzesList quizzes owned by the connected QuizGen account, newest first (default 50 per page; the result includes total and has_more). Requires authorization.
get_quizGet one quiz including its full definition. Requires an authorized account that owns the quiz.
update_quizReplace a quiz's definition (full replacement, not a patch — send the complete new definition). The quiz keeps its URL and responses. Requires an authorized account that owns the quiz.
set_quiz_statusOpen ("live") or close ("closed") a quiz. Closed quizzes stop accepting responses but keep their data.
delete_quizDelete a quiz. Its URL stops working and responses become inaccessible. Confirm with your user before calling this.
get_responsesList a quiz's responses (answers, outcome, score, timestamps), newest first (default 50 per page; the result includes total and has_more). Requires an authorized account that owns the quiz.
get_responseGet a single response by id.
get_quiz_statsGet a quiz's funnel stats: views, started, finished, completion rate, and outcome counts.
set_webhookSet (url) or clear (url: null) the webhook for a quiz. Completed responses POST to the url; with a secret, requests carry an X-QuizGen-Signature HMAC-SHA256 header.
send_quiz_by_emailEmail the quiz to recipients — each gets a unique tracked link so responses tie back to them (see get_sends). Any account with a confirmed email address; capped per month and per 24 hours (Free: 200/mo, 50/day; Plus: 5,000/mo). Confirm…
get_sendsDelivery status for a quiz's email sends: sent / started / completed per recipient, plus reminder counts and opt-outs.
set_follow_upSet (or clear with rule: null) automatic email reminders for recipients who haven't completed the quiz. Stops on completion, opt-out, or max_reminders.
check_accountReturn the connected account's email, plan, and limits. Call this to verify that QuizGen authorization (OAuth or API key) is active. (Formerly named check_api_key; the old name still works.)
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"quizgen": {
"type": "http",
"url": "https://quizgen.dev/mcp"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http quizgen https://quizgen.dev/mcpclaude.ai, ChatGPT and other clients: add https://quizgen.dev/mcp as a custom connector in the settings.
OpenAPI spec
10 paths, read 2026-10-28. Open the spec ↗
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.