langchain-ai · robin-ede · Jul 10, 2025
diff --git a/README.md b/README.md
@@ -177,7 +177,7 @@ The second memory schema we provide is the **Note** schema, shown below:
 }
 ```
 
-Just like the previous example, this schema has a name, description, and parameters. Notic that the `update_mode` this time is "insert". This instructs the LLM in the memory service to **insert new memories to the list or update existing ones**. The number of memories for this `update_mode` is **unbound** since the model can continue to store new notes any time something interesting shows up in the conversation. Each time the service runs, the model can generate multiple schemas, some to update or re-contextualize existing memories, some to document new information. Note that these memory schemas tend to have fewer parameters and are usually most effective if you have a field to let the service provide contextual information (so that if your bot fetches this memory, it isn't taken out-of-context).
+Just like the previous example, this schema has a name, description, and parameters. Notice that the `update_mode` this time is "insert". This instructs the LLM in the memory service to **insert new memories to the list or update existing ones**. The number of memories for this `update_mode` is **unbound** since the model can continue to store new notes any time something interesting shows up in the conversation. Each time the service runs, the model can generate multiple schemas, some to update or re-contextualize existing memories, some to document new information. Note that these memory schemas tend to have fewer parameters and are usually most effective if you have a field to let the service provide contextual information (so that if your bot fetches this memory, it isn't taken out-of-context).
 
 To wrap up this section: `memory_schemas` provide a name, description, and parameters that the LLM populates to store in the database. The `update_mode` controls whether new information should always overwrite an existing memory or whether it should insert new memories (while optionally updating existing ones).
-Original file line number
+Diff line change
@@ Expand Up @@
     }
     ```
-    Just like the previous example, this schema has a name, description, and parameters. Notic that the `update_mode` this time is "insert". This instructs the LLM in the memory service to **insert new memories to the list or update existing ones**. The number of memories for this `update_mode` is **unbound** since the model can continue to store new notes any time something interesting shows up in the conversation. Each time the service runs, the model can generate multiple schemas, some to update or re-contextualize existing memories, some to document new information. Note that these memory schemas tend to have fewer parameters and are usually most effective if you have a field to let the service provide contextual information (so that if your bot fetches this memory, it isn't taken out-of-context).
+    Just like the previous example, this schema has a name, description, and parameters. Notice that the `update_mode` this time is "insert". This instructs the LLM in the memory service to **insert new memories to the list or update existing ones**. The number of memories for this `update_mode` is **unbound** since the model can continue to store new notes any time something interesting shows up in the conversation. Each time the service runs, the model can generate multiple schemas, some to update or re-contextualize existing memories, some to document new information. Note that these memory schemas tend to have fewer parameters and are usually most effective if you have a field to let the service provide contextual information (so that if your bot fetches this memory, it isn't taken out-of-context).
     To wrap up this section: `memory_schemas` provide a name, description, and parameters that the LLM populates to store in the database. The `update_mode` controls whether new information should always overwrite an existing memory or whether it should insert new memories (while optionally updating existing ones).
@@ Expand Down @@