Entelgia — a dialogue-governed cognitive architecture running fully locally (Ollama)

Show and Tell
1

Hi everyone,

I’ve been building an experimental AI system called Entelgia.

Instead of focusing on prompt engineering, the project explores how
dialogue, internal state, memory, and a meta-observer agent can govern
emergent behavior between AI agents.

The system runs fully locally using Ollama and now includes:
• dynamic speaker selection
• observer (Fixy) intervention policy
• formal system specification (SPEC.md)
• automatic installer

I would really appreciate feedback from the community.

1 Like
2

Additional context:

Entelgia is not meant to replace agent frameworks,
but to explore whether dialogue itself can act as a governance layer
over internal cognitive state.

I’m especially interested in feedback about:
• observer-agent design
• dialogue stability
• evaluation signals for emergent behavior

1 Like