MCP Questionnaire
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
MCP Questionnaire
Structured input surfaces inside the Claude conversation — prefilled forms, assumption ledgers, review matrices — instead of clarifying questions as prose bullets. The agent infers what it can, renders a form already prefilled with provenance on every value, and your job is auditing a proposal, not authoring one. Answers flow straight back into model context.
DESIGN.html is the canonical spec (open it in a browser — the mockups are clickable). This README only tells you how to install and run.
Install (Claude Code)
One install gives you the connector and the skill that knows when a form is the right move:
/plugin marketplace add JakubMMazurek/MCP-Questionnaire
/plugin install mcp-questionnaire@mcp-questionnaireor from a shell:
claude plugin marketplace add JakubMMazurek/MCP-Questionnaire
claude plugin install mcp-questionnaire@mcp-questionnaireAt 365 Business Finance, skip both: the plugin arrives by organization sync
from the private mirror, and the connector is added separately by an Owner.
See packages/plugin/README.md.
Related MCP server: chimeralang-mcp
Install (claude.ai / Claude Desktop — where the forms render)
Settings → Connectors → Add custom connector → paste the URL from
packages/plugin/.mcp.json. Claude will send you
to Microsoft to sign in; if you are on the Tech Team Access policy you come
back connected. First render asks a one-time "Always allow" permission. Add on
web/desktop first — mobile picks it up from there.
Access control
The connector URL is public and carries nothing. Every MCP request needs an OAuth token, and the only way to get one is to sign in through Cloudflare Access, which federates to Microsoft Entra ID and admits only members of the Tech Team policy in Zero Trust. Anyone else is stopped at Access's own login page and never reaches this Worker.
Add or remove a teammate: edit the Tech Team rule group in Zero Trust
(Access controls → Policies). No code change, no redeploy — and nothing in this
repo lists who has access, deliberately. A leaver loses access when their Entra
account is disabled, because the policy requires Entra as the login method. A
token someone already holds expires on its own (or
delete their grant:/token: keys from the OAUTH_KV namespace to cut it
short).
This is a deliberate reversal of the original "no auth" decision (DESIGN.html §3): the URL path used to be the capability, which only worked while this repo stayed private, and that stopped being true once the connector had to be installed from a plugin marketplace. Form ids remain capabilities internally.
Repo map
| the spec — every decision, with rationale |
| meta-schema, Zod validator with teaching errors, path parser ( |
| renderer: rules engine, host bridge, single-file bundle ( |
| Cloudflare Worker: MCP handler, form Durable Objects ( |
| the installable plugin (connector + skill) |
| marketplace catalog pointing at |
Develop
pnpm install
pnpm -r test # 498 tests
pnpm --filter @mcpq/ui dev # offline host harness at localhost:5173/dev/
pnpm build # single-file renderer bundle, verified self-containedDeploying (packages/worker): pnpm exec wrangler deploy. Two secrets:
wrangler secret put ACCESS_CLIENT_SECRET
wrangler secret put FORM_URL_SECRETACCESS_CLIENT_SECRET comes from the Access for SaaS OIDC application in
Zero Trust whose Redirect URL is
https://mcp-questionnaire.365finance.workers.dev/callback; it is shown only
at creation, so getting it again means Reset secret. Its client id and
issuer are vars in wrangler.jsonc, not secrets — neither is a credential,
and both are visible in the public discovery document. Pipe secrets with
tr -d '\n': a trailing newline has bitten this deploy before. .dev.vars
holds an obvious fake for local runs.
This server cannot be deployed
Maintenance
Related MCP Connectors
Your audit methodology inside Claude: findings, risks, controls and workpapers in your team format.
Turn PRDs and product ideas into structured specs so coding agents build your intent, not theirs.
A structured financial modeling layer for AI agents. Build, version, and audit financial models without drift, then export to Excel, from Claude or any MCP client. Learn more: https://layerz.cc/for-agents
Live data grids for AI agents. Push structured data; humans review, agents read back via MCP.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to create visual, interactive interfaces with diagrams, structured input forms, and persistent state management instead of text-only responses.7-
- AlicenseCqualityAmaintenanceProvides Claude with 44 tools for confidence gating, typed outputs, hallucination detection, and constraint enforcement during conversations.5144 PyPI1MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to ask clarification questions and receive structured user input through a Human-in-the-Loop interface.1MIT
- FlicenseNot gradedqualityBmaintenanceEnables Claude Code agents to perform structured decision-making through TypeSafe Jev, supporting yes/no checks, route selection, scoring, and batched multi-question judgments with confidence values.1-