PipedreamHQ / PipedreamHQ/pipedream
[FEATURE] NuvemShop / TiendaNube
- Dominant language
- JavaScript
- Stars
- 11.7k
- Forks
- 5.8k
- Avg merge
- 3d 10h
- Merged PRs (30d)
- 102
Description
### **Is your feature request related to a problem? Please describe.**
We are an agency working mainly with LATAM customers on NuvemShop / TiendaNube. Currently, we face a scalability bottleneck when building sales and post-sales workflows for our clients.
To enable AI agents to manage contacts or create "deals" (sales opportunities/draft orders), we currently have to build **deterministic, custom API integrations for each individual customer**. Because every store has slightly different needs or external CRM connections, we are rewriting glue code constantly. This approach is rigid, hard to maintain, and does not scale as we take on more clients.
### **Describe the solution you'd like**
We would like NuvemShop / TiendaNube to implement **Model Context Protocol (MCP)** support.
Specifically, we need an MCP Server implementation that exposes the store's key resources (Products, Customers, Orders, and Draft Orders) to AI agents in a standardized way. This would allow an AI agent to:
1. **Read Context:** Instantly understand a specific store's inventory and customer list without custom hard-coded logic.
2. **Take Action:** autonomous creation of "Deals" (e.g., creating a Draft Order or updating a Customer note) and management of contacts through a standard protocol.
By supporting MCP, you would allow developers to build a single "Sales AI Agent" that works across *any* NuvemShop store immediately, rather than building custom API connectors for every single merchant.
### **Do you have a workaround?**
Yes, but it is not scalable. We are currently manually developing custom API integrations for each customer using the standard REST API (e.g., `https://api.tiendanube.com/v1/...`). This requires maintaining separate codebases and "hard-coding" logic for how the AI interacts with the store's specific data structure, which defeats the purpose of using a flexible AI agent.
### **Comparable features in other tools?**
**Shopify** and other major platforms are seeing rapid adoption of agentic AI tools. The **Model Context Protocol (MCP)** (backed by Anthropic and others) is emerging as the industry standard for connecting LLMs to external data. Adopting this would position NuvemShop as a leader in AI-ready e-commerce infrastructure in LATAM.
### **Additional context**
We specifically need to streamline the "Whole Sales" and "Post-sales" workflows.
* **Current API Documentation we use:** [https://tiendanube.github.io/api-documentation/intro](https://tiendanube.github.io/api-documentation/intro)
* **Desired Workflow:** An AI agent connects to a NuvemShop store via MCP, searches for a customer, checks their order history, and creates a new Draft Order (Deal) based on a chat conversation, all without us writing a single line of custom REST API boilerplate for that specific store.
Contributor guide
Assessment
This issue has not been assessed yet.