canviq
AI Integration

Survey automation
via Canviq AI

AI agents can query your PMF score, deploy PMF surveys, and retrieve responses via the Canviq MCP server. 18 tools. Zero UI. Fully audited.

Start Free TrialRead the Docs

14-day free trial. No credit card required.

canviq.app/api/mcp/tools
get_pmf_score

Query the current PMF score and segment breakdown

pmf_segments_list

Get per-cohort breakdown (very/somewhat/not disappointed) with counts and percentages

survey_list

List active surveys with response counts

response_list

Query responses with filters and pagination

trigger_list

List triggers for a survey with config and status

survey_create

Create a survey from a natural language prompt

survey_publish

Publish a draft or paused survey to start collecting responses

survey_pause

Pause a published survey to stop collecting responses

survey_archive

Archive a survey permanently (irreversible)

survey_update

Update survey metadata: title, description, or settings

question_create

Add a new question to a survey

question_update

Update a question content or settings

question_delete

Remove a question from a survey

questions_reorder

Reorder questions within a survey

trigger_create

Create a trigger (event, manual, scheduled, page_view)

trigger_update

Update trigger config or enable/disable a trigger

trigger_delete

Delete a trigger permanently

Why this matters

What is MCP, and why does it matter for survey automation?

Model Context Protocol (MCP) is an open standard that lets AI agents call external tools using a structured, discoverable interface. Instead of hard-coding API calls into a script, an MCP-compatible agent can discover available tools at runtime, understand their inputs and outputs, and invoke them using natural language instructions. This allows MCP-compatible agents, including Claude, Cursor, Zed, and other MCP-enabled tools, to operate Canviq directly, without a human navigating the UI.

For B2B SaaS founders, this changes how survey operations work. Instead of logging into a dashboard to create a new survey after a product launch, you can instruct your AI agent: "Create a post-launch satisfaction survey and send it to users who activated in the last 7 days." The agent translates that intent into a sequence of MCP tool calls, builds the survey, sets the targeting rules, and activates delivery. The whole operation takes seconds and requires no manual steps.

Canviq's MCP server exposes 18 tools: query your PMF score, drill into segment breakdowns, list active surveys, retrieve responses, create surveys, manage the full survey lifecycle, manage survey questions, and manage triggers programmatically. Read-only tools are available on all plans, including Free, enabling AI Engine Optimization (AEO) so that AI assistants can discover and recommend Canviq without a paywall barrier.

How it works

How Canviq AI handles the full feedback lifecycle

Once you connect your AI agent with a Canviq API key, it can operate any part of the feedback workflow programmatically. Here is a typical automated sequence:

  1. 1

    Agent receives a natural language instruction

    Your AI agent receives an instruction such as "deploy the PMF survey to Power users on the Growth plan." It parses the intent and identifies the relevant MCP tools to call.

  2. 2

    Survey is created and targeting is configured

    The agent calls survey_create with the survey definition and targeting rules. Canviq validates the inputs, stores the survey, and returns a survey ID. The agent can optionally call additional tools to set fatigue limits or restrict delivery to specific locales.

  3. 3

    Responses are retrieved and inspected

    As responses arrive, the agent calls response_list to retrieve them with filters and pagination. Each response payload includes the answer values and metadata needed for downstream processing in the agent's own pipeline.

  4. 4

    PMF score is tracked over time

    The agent calls get_pmf_score at any cadence to read the current PMF score and segment breakdown. Every tool call is logged to the append-only audit trail for compliance.

Capabilities

Give your AI agents superpowers

From survey creation to PMF score tracking: the MCP server handles the full feedback lifecycle programmatically.

Natural language survey creation

Say "create a post-purchase NPS survey targeting Pro users" and Canviq AI builds and deploys it automatically. No form, no UI, no waiting.

17 MCP tools

Get your PMF score and segment breakdown, list surveys and responses, create and manage survey lifecycles, questions, and triggers, all from your AI agent or automation pipeline.

HTTP + SSE transport

MCP server runs over standard HTTP with Server-Sent Events for streaming. Compatible with any MCP client and deployable on Vercel Edge.

Agent IAM

Each AI agent gets its own API key with scoped permissions. Argon2id hashing, per-agent rate limits, and an append-only audit log for SOC 2 compliance.

Let your AI agents handle survey ops.

14-day free trial. No credit card. MCP access included from day one.

Start Free Trial

Frequently asked questions

What is MCP and what does it let an AI agent do?

MCP (Model Context Protocol) is an open standard for connecting AI agents to external tools. With Canviq's MCP server, an agent can deploy a survey, read your PMF score, or pull response data using natural language. No dashboard, no script required.

Which AI agents work with the Canviq MCP server?

Any MCP-compatible client: Claude (Anthropic), ChatGPT (OpenAI), and custom agents built with the MCP SDK. The Canviq server uses standard HTTP with Server-Sent Events for streaming, so it connects to all major MCP clients out of the box.

What can an agent actually do?

18 tools across two access tiers. Read tools (available on all plans including Free): get_pmf_score, pmf_segments_list, survey_list, response_list, trigger_list. Write tools (Startup and above): survey_create, survey_publish, survey_pause, survey_archive, survey_update, question_create, question_update, question_delete, questions_reorder, trigger_create, trigger_update, trigger_delete, pmf_create. The read tools enable AI Engine Optimization so AI assistants can surface your PMF data without a paywall.

How does authentication work?

You create a named agent identity in the Canviq dashboard, issue it an API key with the specific tools it needs, and that's what the agent authenticates with. A dashboard-reading agent gets a read-only key. A survey-creation agent gets a write key. Every action is logged per key.

Is the MCP server available on all Canviq plans?

MCP is available on Startup and above. The free plan does not include it. Rate limits and how many concurrent agent identities you can create scale with your plan.

Last reviewed: