Add more robust parsing of LLM JSON output

agent/nodes/verificationSetup.ts

@@ -1,8 +1,7 @@
 import { GraphNode } from "@langchain/langgraph";
 import { MessagesState, ProposedTriggerEventArray } from "../state";
 import { logger } from "../utils/logger";
-import { queryScraper } from "../tools/webSearch";
-import { rankAndDisplayData } from "../tools/triggerEventTools";
+import { jsonrepair } from 'jsonrepair'
 
 export const verificationSetup: GraphNode<typeof MessagesState> = async (state) => {
   //this is kinda doing two things, but having two nodes for it seems overkill
@@ -11,15 +10,29 @@ export const verificationSetup: GraphNode<typeof MessagesState> = async (state)
     logger.warn("No trigger events in memory, parsing")
 
     let genResponse = state.messages.at(-1)?.content.toString() ?? "";
-    const parsed = ProposedTriggerEventArray.parse(JSON.parse(genResponse));
 
-    for (let i = 0; i < parsed.length; i++) {
-      const search = parsed[i].SearchQuery
-      // const data = await queryScraper(search);
-      // const output = await rankAndDisplayData(data, search);
+    const repaired = jsonrepair(genResponse);
 
-      // parsed[i].context = output;
-      parsed[i].context = "NONE"
+    let parsed;
+
+    try {
+      const json = JSON.parse(repaired);
+
+      if (Array.isArray(json)) {
+        parsed = ProposedTriggerEventArray.parse(json);
+      } else {
+        // try grab first value
+        const firstValue = Object.values(json)[0];
+
+        if (Array.isArray(firstValue)) {
+          parsed = ProposedTriggerEventArray.parse(firstValue);
+        } else {
+          throw new Error("No array found in JSON");
+        }
+      }
+    } catch (err: any) {
+      logger.error(`Failed to parse LLM response: ${err.message}`);
+      throw new Error(`Failed to parse LLM response: ${err}`);
     }
     
     return { proposedTriggerEvent: parsed, proposedTriggerEventIndex: 0 };

agent/package-lock.json

@@ -20,6 +20,7 @@
         "dotenv": "^17.2.3",
         "exponential-backoff": "^3.1.3",
         "fs": "^0.0.1-security",
+        "jsonrepair": "^3.13.3",
         "langchain": "^1.2.14",
         "selenium-webdriver": "^4.40.0",
         "tldts": "^7.0.23",
@@ -2075,6 +2076,15 @@
       "integrity": "sha512-ZClg6AaYvamvYEE82d3Iyd3vSSIjQ+odgjaTzRuO3s7toCdFKczob2i0zCh7JE8kWn17yvAWhUVxvqGwUalsRA==",
       "license": "ISC"
     },
+    "node_modules/jsonrepair": {
+      "version": "3.13.3",
+      "resolved": "https://registry.npmjs.org/jsonrepair/-/jsonrepair-3.13.3.tgz",
+      "integrity": "sha512-BTznj0owIt2CBAH/LTo7+1I5pMvl1e1033LRl/HUowlZmJOIhzC0zbX5bxMngLkfT4WnzPP26QnW5wMr2g9tsQ==",
+      "license": "ISC",
+      "bin": {
+        "jsonrepair": "bin/cli.js"
+      }
+    },
     "node_modules/jszip": {
       "version": "3.10.1",
       "resolved": "https://registry.npmjs.org/jszip/-/jszip-3.10.1.tgz",

agent/package.json

@@ -24,6 +24,7 @@
     "dotenv": "^17.2.3",
     "exponential-backoff": "^3.1.3",
     "fs": "^0.0.1-security",
+    "jsonrepair": "^3.13.3",
     "langchain": "^1.2.14",
     "selenium-webdriver": "^4.40.0",
     "tldts": "^7.0.23",