standardize inner thoughts terminology

letta-ai / letta

Letta (formerly MemGPT) is a framework for creating LLM services with memory.

https://letta.com

Apache License 2.0

12.22k stars 1.35k forks source link

standardize inner thoughts terminology #1918

Open kl2806 opened 2 weeks ago

kl2806 commented 2 weeks ago

Currently, there are several terms that refer to the same concept in the repo:

inner thoughts
inner monologue
internal monologue To avoid confusion, these should be standardized to a single name eg. "inner thoughts," and we should only have different terminology if we add additional conceptual differences i.e. thoughts that are only visible to a single agent in a mult-agent system.

raolak commented 2 weeks ago

The current stream response should have more well defined spec . Also leverage opeai feature so that clients can building much more flexible rendering capabilities .

Example Based on type of response (md, csv, chart, code, image, diff, data payloads etc ), one should able to design rendering plugin components... even its possible the executable plugins can be part of stream response. The UI execution sandbox within the browser can safely execute those executable plugins