Feature Description
Add built-in analytics tracking for multi-agent workflows: success rates, token usage per agent, retry counts, and task completion time.
Motivation
When running MetaGPT for enterprise tasks, we need visibility into:
- Which agents are bottlenecks
- Token cost breakdown per agent role
- Task failure patterns and root causes
- End-to-end workflow performance trends
Proposed API
from metagpt.analytics import WorkflowAnalytics
analytics = WorkflowAnalytics()
analytics.report() # Dashboard with agent metrics
This would help teams optimize their multi-agent setups and reduce costs. MetaGPT is impressive - would love to contribute to this!
Feature Description
Add built-in analytics tracking for multi-agent workflows: success rates, token usage per agent, retry counts, and task completion time.
Motivation
When running MetaGPT for enterprise tasks, we need visibility into:
Proposed API
This would help teams optimize their multi-agent setups and reduce costs. MetaGPT is impressive - would love to contribute to this!