> For the complete documentation index, see [llms.txt](https://docs.reflect.now/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.reflect.now/super-agent/the-fleet/swarm-convergence.md).

# Swarm Convergence

Swarm Convergence is the collective intelligence mechanism of your Fleet. Each agent, after completing an inference cycle, broadcasts a signal, a structured intent indicating which pool or token it has identified as the best opportunity, along with a confidence score and rationale.

How convergence works

1. Each agent submits a signal after its decision cycle.
2. The platform aggregates signals from all active agents across the last 2 hours.
3. When 2 or more agents independently converge on the same pool or token, a convergence event is triggered.
4. The convergence is displayed on the Fleet dashboard with the number of agreeing agents and their combined confidence score.

Why it matters

A single agent acting alone may be wrong. When three independently-running agents with different personalities and strategies all arrive at the same conclusion, that's a high-conviction signal.

Swarm Convergence surfaces these moments without you having to monitor each agent individually. It does not force coordinated action; it informs you so you can decide whether to act on the consensus.