Querying the Vector embeddings stored by Semantic Recall
I am using Mastra's Memory (https://mastra.ai/docs/memory/semantic-recall) to store the agent context and tool call output to the vector db.
I was hoping to use this https://mastra.ai/docs/reference/tools/vector-query-tool to make the agent manually query the embeddings stored by Semantic recall too, instead of only querying the first time with the initial user prompt.
What should I pass in as indexName for this to work?? Help would be appreciated 🙏🙏
Semantic Recall | Mastra Documentation
Learn how to use semantic recall in Mastra to retrieve relevant messages from past conversations using vector search and embeddings.
createVectorQueryTool() | Mastra Documentation
Documentation for the Vector Query Tool in Mastra, which facilitates semantic search over vector stores with filtering and reranking capabilities.

4 Replies
I was snooping around the db and it seems like it's storing the embeddings memory_messages_1024 table
what does indexName actually represent? in the db tables?
📝 Created GitHub issue: https://github.com/mastra-ai/mastra/issues/9390
GitHub
[DISCORD:1432632540442792047] Querying the Vector embeddings stored...
This issue was created from Discord post: https://discord.com/channels/1309558646228779139/1432632540442792047 I am using Mastra's Memory (https://mastra.ai/docs/memory/semantic-recall) to stor...
indexName is the name of the vector index, which in this case is the name of the "table", memory_messages_1024 😉oh alr8, thanks man