How the Particl Connector handles your data, authentication, and privacy.
TL;DR
If you're considering connecting Particl to ChatGPT, Claude, Cursor, or another AI tool, the most important question is usually the simplest one:
Can Particl see my data?
No. The connector does not read your conversations, access your workspace, or collect your internal business data. It only lets your AI tool request Particl's market research data on your behalf.
No. The Particl Connector is a read-only data source. Its job is to return Particl market intelligence data when your AI tool asks for it. That includes things like competitor products, pricing, sales estimates, and marketing assets.
There is no mechanism for the connector to read your chats, prompts, files, or documents inside ChatGPT, Claude, Cursor, or any other AI workspace.
The flow is simple: You ask your AI tool a question → the AI tool calls Particl → Particl returns market data → the AI tool answers you.
No. The connector does not connect to your Shopify store, ad platforms, customer lists, analytics tools, or any other internal systems. It has zero integration with your proprietary data sources.
The data available through the connector is Particl's own market research dataset — the same kind of competitor intelligence you already access in the Particl dashboard. That includes public market data such as product catalogs, pricing, estimated sales, emails, and ads.
Only Particl data flows through it.
When you connect the Particl Connector, your AI tool can make requests like:
All of the connector's tools are read-only. There are no write operations. It cannot modify anything in your account, in your AI tool, or in any other system.
No. When you connect using OAuth, your token is short-lived, stateless, and expires automatically — it is never stored on Particl's servers. If you connect using an API key, the key is cached in server memory briefly (up to 5 minutes) for performance, but it is never written to disk or a database. In both cases, your credentials are not persisted.
The connector is designed to be stateless, which means it does not keep:
Each request stands on its own.
Only what's needed for billing and usage tracking. When you use a connector tool, Particl records which tool was called (e.g. "search_companies"), the number of results returned, the credit cost, and the search parameters you passed (such as a company name or product type filter). This is the same kind of usage data tracked when you export a CSV from the dashboard.
Particl does not log your conversations, prompts, AI-generated responses, or any content from your AI workspace.
The connector uses MCP (Model Context Protocol), an open standard for connecting AI tools to external data sources. Through MCP, your AI assistant can call Particl as a data source when you ask a market research question.
A simple way to think about it:
You → AI Tool → Particl Connector → Particl API → market data returned → AI Tool → You
The important part is this: Data flows from Particl to your AI tool — not from your AI tool back into Particl.
You stay in control.
Connector usage draws from your existing Particl export credit pool, just like CSV exports.
Yes. The Particl Connector is built so your AI tool can query Particl's market intelligence, not so Particl can inspect your workspace. It is read-only, stateless, and limited to returning Particl data on authenticated requests. It does not access your chats, your files, your store, your ad accounts, or your internal systems.
If you still have questions about your setup, contact support@particl.com and we'll be happy to help.