A lot shipped this month across agents, workflows, and how teams actually use Scout day to day. From better Python support to AI visibility, this one’s about making performance issues easier to find and faster to act on.
Python agent
- Redis Cluster instrumentation support
- Starlette 1.0 compatibility
- Python 3.13 and 3.14 support added; 3.8 and 3.9 dropped (end-of-life)
- 401 responses no longer tagged as errors
- RQ
Job.idcompatibility fix afterget_id()was removed upstream
Ruby 6.2.0
- A fix for HTTP instrumentation with http gem v6.0 and above.
- Automatically supports both v5 and v6 without any config changes
Elixir 2.0
- A major version release for the Elixir agent. If you’re on Phoenix, it’s worth the upgrade.
- Stability improvements including AgentManager OOM fixes for long-running applications.
Scout CLI v0.3
- TOON output format for token-efficient LLM consumption, auto-enables when piped
- New
scout usagecommand to track transactions by day or app. Set-up docs here.
MCP server
- Added background job monitoring with throughput, execution time, and latency metrics
- Usage visibility now available inside your AI assistant
Scout is now on AWS Marketplace: If your team procures software through AWS, this is the easiest path to getting Scout on your account. View on AWS Marketplace.
From the Scout Community: BackerKit
BackerKit manages $3.7 billion in crowdfunding pledges with five engineers. They switched from Datadog to Scout and the biggest change was how their team spends time. Their performance reviews went from hunting to a prioritized list of things to fix. They also have Scout MCP wired into GitHub Copilot for trace investigation. Read the full story.
Events: Where we’ll be
Blue Ridge Ruby — April 30 and May 1, Asheville NC Ruby conference in the mountains. If you are going, come say hi.
JetBrains PHPverse 2026 — June 9th, online A free streaming PHP event with 55,000+ developers worldwide. Scout is a proud sponsor. Grab your spot!
Coming soon
- Anomaly detection — automatically flags unusual behavior in your endpoints. Tag critical endpoints you want to keep a close eye on.
- One-click issue creation — open tickets in GitHub, Linear, or Jira with context pre-filled when Scout surfaces a problem.
One more thing
If your team is using AI tools day to day, we would love 20 to 30 minutes to hear how you are using them and where the gaps in visibility are. Grab time here if you are interested.
Happy Scouting,
Sarah Product @ Scout Monitoring