From the source
From the source
Cognition argues against multi-agent architectures for building AI agents, advocating instead for single-threaded linear agents with context compression.
The principles of context engineering include sharing full agent traces and avoiding conflicting implicit decisions.
From the source
I would argue that Principles 1 & 2 are so critical, and so rarely worth violating, that you should by default rule out any agent architectures that don’t abide by them. You might think this is constraining, but there is actually a wide space of different architectures you could still explore for your agent. The simplest way to follow the principles is to just use a single-threaded linear agent:
cognition.com
