this will allow skill makers and end-users to consider LILACS a black box and still get its answers and configure functionality
this skill should:
have a crawler instance #9
have a question parser instance #21
methods/intents for default questions #15
tutorial / info about LILACS
intents to vocally configure some LILACS variables
intents to vocally output LILACS related info
listen on messagebus for all kinds of LILACS related requests <- answer questions /diagnostics info
keep only relevant nodes on memory
manage the crawler
interact with concept storage (function of the crawler) #1
handle the learning in case of intent failure #18
maybe start the knowledge and storage services so we dont need more processes in line with #20
we must decide when we want this skill to act, i believe it should be in case of intent failure after wolfram alpha
having intents for properly formatted questions should be enough for normal use, but we also want it to output something to user and gather info when he doesnt know anything,
A central skill should be the core of LILACS
this will allow skill makers and end-users to consider LILACS a black box and still get its answers and configure functionality
this skill should:
we must decide when we want this skill to act, i believe it should be in case of intent failure after wolfram alpha
having intents for properly formatted questions should be enough for normal use, but we also want it to output something to user and gather info when he doesnt know anything,