Content
@
https://warpcast.com/~/channel/kevinoconnell
0 reply
0 recast
0 reaction
kevin
@kevinoconnell
https://x.com/8baller11/status/1889729066636341652 @oconnell and I have been working alongside 8baller, a top ecosystem developer on Olas (haven't gotten him on here yet 😀) to build an abstraction library around it called Auto Dev tl;dr: it abstracts away a lot of the boilerplate command logic that you would have had to do otherwise. now you can 1.) create an agent 2.) fork components from an agent 3.) run an agent 4.) run a service (via docker compose) 5.) run entire workflows within a yaml file you still need to write business logic, but this removes away 98% of the annoying parts that developers would run into before really excited about this, i think it will be a lot easier for people to make novel use cases with this now
2 replies
1 recast
8 reactions
kevin
@kevinoconnell
why use Olas? Olas the most robust generalized on chain agent framework - enabling use cases from automated prediction market agents, to yield optimizers, to character agents, amongst other examples. lfg!
1 reply
0 recast
2 reactions
shoni.eth
@alexpaden
is it built to handle (commands/actions/plugins) and if so are they triggered by an LLM or by regex? a lot of us are looking to migrate off elizaOS
1 reply
0 recast
1 reaction
kevin
@kevinoconnell
yeah olas is made for autonomous workflows whereas eliza is literally just for character agents, not workflows. by default an olas agent has no llm functionality, although that can be added as part of a skill (like a LLM decision can be a step of the process, etc)
1 reply
0 recast
1 reaction
shoni.eth
@alexpaden
would be interested in getting more information on workflows, i haven't found either topic particularly compelling besides a social agent that summarizes new/relevant research for /aichannel not clueless, just haven't needed anything i guess
1 reply
0 recast
1 reaction
