Is it possible to upload documents to a LLM via clay from a webhook? We are looking to use Clay for some of our operations as well and have a use case to analyze documents via a LLM which we are currently doing manually. Appreciate any feedback.
Yes, you can send data to Clay via webhooks and then process it with LLM integrations. Clay accepts HTTP POST requests in JSON format through webhook endpoints, allowing you to automate data imports from external sources. You can set up a webhook table by clicking "+ Add" in your workbook, searching for "Webhooks, " and selecting "Monitor webhook" to get your unique URL. Once document data arrives via webhook, you can use Clay's enrichment columns (including LLM integrations) to analyze the content automatically. After processing, you can push results back to your system using our HTTP API integration. This creates an automated workflow: webhook receives documents → Clay processes with LLM → results sent back to your application.
Jason, you can use HTTP API columns to send data elsewhere, assuming those services have API endpoints you can work with for the purpose. If you need Clay to fetch data from elsewhere, you need only provide the link/URL in your prompt, which the LLM can review, process, and take necessary action on. We're still building out the ability to upload docs as context in Use AI columns. That will be manual (e.g. important internal business documents on company, messaging, ICP, etc.) rather than dynamic (row-by-row context), but keep an eye out! More to come.
This thread was picked up by our in-app web widget and will no longer sync to Slack. If you are the original poster, you can continue this conversation by logging into https://app.clay.com and clicking "Support" in the sidebar. If you're not the original poster and require help from support, please post in 02 Support.