Build AI Agents on Holded

Build intelligent AI Agents on top of Holded with Peliqan, the leading data foundation for the Agentic AI world.

Build AI Agents

Build AI Agents on top of Holded

Build intelligent AI Agents on top of Holded, with Peliqan’s AI data foundation:

  • Holded MCP Server
  • Holded in n8n AI Agents
  • Holded in Make
  • “Text to SQL” on Holded data
  • Holded RAG with out-of-the-box embeddings (vectors)
  • Holded Graph RAG
  • Query unified 360° data combining Holded and other data

Holded MCP Server

Publish a Holded MCP Server to query data from Holded and to take actions in Holded such as doing updates and adding new data in Holded.

Build MCP Server

Build AI agents in n8n

Holded in n8n

Build AI Chatbots and AI Agents in n8n that can perform “Text to SQL” to query Holded data and perform RAG and Graph RAG on information from Holded.

Implement a “Text to SQL” chatbot on Holded

Implement an AI Chatbot that can answer analytical data questions on Holded data using “Text to SQL”. 

Text to SQL

Implement a chatbot

Implement a chatbot with RAG on Holded

Implement RAG (retrieval-augmented generation) on Holded data with an out-of-the-box vector store (embeddings) of all business entities and other information in Holded.

Combine Holded data with 250+ sources

Combine data from Holded with data from 250+ other connectors, and build 360° views of business entities such as customers, leads, products, employees etc.

Feed unified 360° data models to your AI Agents with RAG and “Text to SQL”. Allow your AI Agents to access all business data in one uniform data model.

Combine data

Prepare your Holded data for AI

Access, combine, and report on data from Holded and all your SaaS apps instantly. Gain valuable insights by bringing all your business data together in one place within minutes.

SaaS Data Cockpit

Unify, Automate & Activate Your Data 

Connect all your SaaS apps, databases, and spreadsheets into one workspace. Build automations, analytics pipelines, and data apps — all in one place.

G2 Badges

Frequently Asked Questions

Why do I need Peliqan for my AI ?

Peliqan is an all-in-one data platform with 250+ data connectors (ERP, CRM, Accounting, ATS/HRM, cloud storage etc.) – including Holded – and a built-in data warehouse. Peliqan allows you to unleash, prepare and combine your Holded data for AI, including relational & non-relational data. Peliqan turns your Holded data into 360° views that can be used in AI Agents built in n8n, Make, langChain, langGraph or any other framework. Use Peliqan to create embeddings, store them in a vector store so that your AI chatbots can use RAG and Graph RAG, combined with Text-to-SQL for analytical reasoning. Peliqan is the only platform that allows your AI Agents to combine RAG and Text-to-SQL to apply deep reasoning on your Holded data. Use Peliqan to expose any Holded as an MCP server to query data and to take actions.

There are different ways to build an AI agent that can query data in Holded and take actions in Holded. For example you can build an AI agent in n8n and use Peliqan as the data foundation. Peliqan will sync your Holded data to its built-in data warehouse and allow the AI Agent to perform “Text to SQL” and RAG to answer questions and to perform reasoning on Holded data, combined with data from 250+ other sources.

First sign up for a free trial on Peliqan.io, next connect Holded in Peliqan. Once that is done, create an AI agent in n8n and use the Peliqan n8n node in your worflow. Add Peliqan as a “tool” to your AI Agent node, so that the AI agent can query your Holded data using Text to SQL.

There are different options to use RAG (retrieval augmented generation) in your AI Agent with Holded data. One option is to create a workflow in n8n that fetches all Holded data from Peliqan and stores it in Supabase as a vector store, with embeddings created using e.g. OpenAI.

In Peliqan, you can set up API endpoints and expose them as MCP Server. In the API endpoint handler script, you can configure actions to be taken in Holded such as querying data, doing lookups, adding new items or performing updates.

n8n is a great tool to build AI chatbots that use Text to SQL, to answer any analytical question on your Holded data. Any question will be converted by the AI agent into an SQL query, which is executed by Peliqan on the Holded data in the data warehouse.

In order to prepare your Holded data for RAG, you need to create embeddings and store them in a vector store. This can be done by creating a workflow in n8n that fetches all Holded data from Peliqan and stores it in Supabase as a vector store, with embeddings created using e.g. OpenAI.

Peliqan data platform

All-in-one Data Platform

Built-in data warehouse, superior data activation capabilities, and AI-powered development assistance.

Peliqan is secure

 Here’s why founders, CIOs, and their IT teams trust us with their data.

SOC 2 Type II

SOC 2 Type 2 validates our security controls, ensuring your data is protected by independently audited security measures.

ISO 27001

ISO 27001 certification (in progress) ensures enterprise-grade information security, protecting your business with globally recognized standards.

GDPR

GDPR compliance guarantees EU data protection compliance, keeping PII data secure within EU boundaries.

Ready to get instant access to all your company data ?