I've spent 8 months refining the foundation of @nineninesai
A self improving agent. Can learn ANY task, build any skill, use any tool.
Best of all, it constantly gets better an cheaper. The same task tomorrow is done more efficiently
StephenAfamO
@StephenAfamO
Founder @nineninesai
- Because @ninenines is built for teams, a lot of thought is built into how permissions are handled. Naturally, Nine should never have access (or memory) about a private chat in a public setting On the other hand, a team wide connection should be usable even in a private chat
- My opinion is that skills are the wrong direction and we would all move away from them in ~1 yearquestion from the TL: > quick question, do you think consuming more skills in the same session unnecessarily dirties the context? > > Say you have a 5 step work, each step with a skill. You can either run all on single session, divide into subagents with relevant skills, or
- The best approach will continue to change and evolve, today it's a single agent, next month its sub agents, 3 months and maybe parallel agents, then a swarm, then maybe back to one agent. No one knows the future, or what new LLM breakthroughs will come.Anthropic is now recommending that companies REMOVE unnecessary subagents from their AI systems. This is a pretty significant shift in agent architecture, with their team finding that a single agent equipped with skills can frequently outperform complex multi-agent systems, and
- Nine will connect to EVERYTHING! If it has an OpenAPI spec, you can connect @nineninesai to it in a few clicks.
