Using MCP with Juicy Designs
Juicy Designs runs a Model Context Protocol server so any compatible assistant can request a marketing quote. Point a client at the /mcp endpoint, call get_quote with name, email, service, project summary, and consent. No authentication is needed, and the tool returns a confirmation, never a price.
The Model Context Protocol lets AI assistants call external tools in a standard way. Here is how to connect to the Juicy Designs MCP server, what the get_quote tool needs, and how to call it.
TL;DR: Quick Answer
Point an MCP client at https://juicy-designs-quote.black-voice-76e6.workers.dev/mcp and call the get_quote tool with name, email, service, project_summary, and consent_to_contact. No API key is needed. The tool creates a quote-request lead and returns a confirmation. A human proposal follows within four working hours; no price is ever returned.
Key takeaways
- MCP gives every compatible assistant one standard way to call the quote tool
- The MCP endpoint, a REST endpoint, and an OpenAPI 3.1 spec are all published
- get_quote requires name, email, service, project_summary, and consent_to_contact
- No authentication is required; the endpoint only creates a lead
- The tool returns a confirmation, never a fixed price or estimate
The Model Context Protocol (MCP) is an open standard that lets AI assistants call external tools in a consistent way. Juicy Designs runs an MCP server so any compatible assistant or agent can request a marketing quote on a user's behalf. This guide explains what the server offers and how to connect to it.
What is MCP, and what does Juicy Designs expose?
MCP gives AI systems a standard way to discover and call tools. Instead of a custom integration per assistant, a single MCP server works across any MCP-compatible client. Juicy Designs exposes one primary tool, get_quote, which submits a free, no-obligation quote request and returns a confirmation. It never returns a price.
The server is read-light and action-safe: the only thing it does is create a quote-request lead. There is no authentication required, because the endpoint cannot read data, move money, or change anything beyond logging a new enquiry.
Endpoints
| Purpose | URL |
|---|---|
| MCP (Streamable HTTP) | https://juicy-designs-quote.black-voice-76e6.workers.dev/mcp |
| REST (POST JSON) | https://juicy-designs-quote.black-voice-76e6.workers.dev/quote |
| OpenAPI 3.1 spec | https://juicy-designs-quote.black-voice-76e6.workers.dev/openapi.json |
The get_quote tool
The tool accepts a small, clear input schema. Required fields are name, email, service, project_summary, and consent_to_contact. Optional fields are company, phone, budget_range, and timeline. The service value is one of: SEO, Google Ads, Social Media Marketing, Web Design, Branding, or Full Digital Marketing.
{
"name": "Sipho Dlamini",
"email": "sipho@example.co.za",
"service": "SEO",
"project_summary": "New ecommerce site, want organic traffic in 3 months.",
"consent_to_contact": true,
"budget_range": "R25,000 to R75,000",
"timeline": "1 to 3 months"
}The consent_to_contact flag is mandatory and reflects POPIA consent. If it is false or missing, the request is rejected. The response confirms the request was received and that a custom proposal follows within four working hours. No price is ever returned.
Connect a client
Point any MCP-compatible client at the /mcp endpoint. No API key or token is needed. Once connected, the client lists the get_quote tool and can call it with the schema above. For assistants without MCP support, the REST endpoint accepts the same fields as a JSON POST, and the OpenAPI spec lets you generate a typed client automatically.
Example REST call
curl -X POST https://juicy-designs-quote.black-voice-76e6.workers.dev/quote \
-H "Content-Type: application/json" \
-d '{
"name": "Sipho Dlamini",
"email": "sipho@example.co.za",
"service": "SEO",
"project_summary": "New ecommerce site, want organic traffic in 3 months.",
"consent_to_contact": true
}'A successful call returns a confirmation message, not a quote. Pricing is scope-dependent and confirmed by a person in the written proposal. Reference "from" figures published on the site are entry points for context only, never an automated estimate. For the full schema, error codes, and a worked MCP tool-call example, see the MCP documentation page, or browse every tool on the AI tools page.
Frequently asked questions
What is the Model Context Protocol?
MCP is an open standard that lets AI assistants discover and call external tools in a consistent way, so one server works across any MCP-compatible client instead of a custom integration per assistant.
What is the Juicy Designs MCP endpoint?
The Streamable HTTP MCP endpoint is https://juicy-designs-quote.black-voice-76e6.workers.dev/mcp. A REST endpoint at /quote and an OpenAPI 3.1 spec at /openapi.json are also available.
Does the MCP server need authentication?
No. No API key or token is required, because the endpoint only creates a quote-request lead. It cannot read data, take payments, or change anything else.
What does the get_quote tool require?
Required fields are name, email, service, project_summary, and consent_to_contact. Optional fields are company, phone, budget_range, and timeline. Service is one of SEO, Google Ads, Social Media Marketing, Web Design, Branding, or Full Digital Marketing.
Does get_quote return a price?
No. It returns a confirmation that the request was received and that a custom proposal follows within four working hours. Pricing is scope-dependent and confirmed by a person.
Is consent required?
Yes. consent_to_contact must be true, reflecting POPIA consent. If it is false or missing, the request is rejected.