Create a Sentry release via sentry-cli or the Releases API with a unique version identifier tied to the deployment artifact
Attach the release name to the SDK configuration (Sentry.init({ release: process.env.SENTRY_RELEASE })) so transactions and errors are tagged
Navigate to Performance > Releases in the Sentry UI to view adoption rate, crash-free session rate, and transaction duration regressions per release
Set up performance alerts on a transaction with a threshold on p75 duration or Apdex score that triggers when a new release degrades beyond baseline
Use the Regression Detection feature to automatically compare a new release to the previous one and surface statistically significant latency regressions
Known gotchas
Release performance comparison requires both releases to have sufficient transaction volume; comparing a canary release (1% traffic) to 100% traffic previous release is statistically unreliable
The release must be associated with commits (sentry-cli releases set-commits) for source code integration and suspect commit identification to work
Performance issue detection requires enabling the performance monitoring product and accepting associated event quota; it is not covered by error-only plans
Give your agent this knowledge — and 200+ more routes
One MCP install gives any agent live access to the full route map, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp