Roadmap
What we’re building next. No dates — just direction.
Recently shipped
Proactive trend alerts
Cost, latency, and success rate drift detection.
Alert grouping
Multiple agents affected by the same issue shown as one alert.
Multi-agent trace linking
Connect child agent runs to their parent pipeline.
Prompt version tagging
Compare cost and quality across prompt versions.
Agent Health Score
One number (0–100) combining all agent signals.
Framework integrations
LangChain, CrewAI, and Vercel AI SDK examples in the docs.
Coming next
OpenTelemetry compatibility bridge
Auto-propagation of trace context across agent boundaries
Alert grouping by root cause
Not just time proximity.
Cost optimization insights
Identify redundant tool calls.
Prompt version filter in runs table
LangChain / CrewAI first-class SDK integration
Beyond documentation examples.
Later
DAG visualization for multi-agent pipelines
Evals system
Structured output quality beyond thumbs up/down.
Team features for Studio plan
Audit log export for compliance
Auto-failover to cheaper model when budget is hit