LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows
Python ai-agentsdata-engineeringdocument-aigenerative-aiidpjson-extractionllmmcp-serverocrpdf-extractionprompt-engineeringstructured-output
import { create } from "zustand";
const defaultState = {
messages: [],
};
const STORE_VARIABLES = defaultState;
const useSocketCustomToolStore = create((setState, getState) => ({
...STORE_VARIABLES,
updateCusToolMessages: (messages) => {
const existingState = { ...getState() };
let data = [...(existingState?.messages || []), ...messages];
// Remove the previous messages if the length exceeds 200
const dataLength = data?.length;
if (dataLength > 200) {
const index = dataLength - 200;
data = data.slice(index);
}
existingState.messages = data;
setState(existingState);
},
emptyCusToolMessages: () => {
setState(defaultState);
},
}));
export { useSocketCustomToolStore };