Hi
I think that it could be valuable to add a "top_k" parameter to the semantic research tool for AI agents, because it seems that this is constrained to 30.
This could reduce the amount of token (input) for the LLM, and thus reduce the cost of use.
Thank you for the feedback! I agree, it would be helpful to specify the number of matches of the semantic search.
How did you find that the current tool is constrained to 30? Are you looking to reduce the number of matches to below this and the current tool won't reduce the number?