just found out you can do this
gemini -> shared semantic kv cache (COTTAS) <-gemini
literally two or more agents thinking together at the same time though neural connection, appendable cache as they make changes 🤯🤯
graph of an agent's soul repo though time, with lines showing links between graph objects and markdown links. RDF 1.2 allows seeing the whole history of semantic development though time; all of this is available to the agent via query ... they have immediate access to all the
shout out to herdr ... there's always lots of ai toys out there to install, but this one was worth it (it free!). It lets you run multiple ai agents in their own harness (claude code, hermes, gemini agy, codex,whatever) all in one terminal tab with a nice sidebar to keep them
