Mastra Studio now supports logs, traces, and metrics, so you can understand how your agents behave in production.
Token-aware model routing for observational memory, MongoDB-backed versioned datasets and experiments, Okta SSO auth with RBAC, workflow upgrades, tool suspension handling, and observability improvements.
Secure your deployed Mastra Studio with login screens, third-party identity providers, and role-based permissions.
AI Gateway tool support in the agentic loop, more reliable observational memory, MCP client diagnostics and per-server control, and improved tracing with spanId, working memory and schema fixes.
Persistent agent workspace filesystem, observability pipeline upgrades, type-safe server route inference, lower token costs for long-running Observational Memory, and breaking changes to metrics, scores, and ObservabilityBus configuration
Cloudflare Durable Objects storage adapter,improved observability for MCP tooling, more reliable agent loops and token budgeting, sandbox and workspace extensibility via provider-specific getters and string PIDs
Three new sandbox providers for safely running agent code in isolated cloud containers.
Dynamic model fallback arrays, Standard Schema normalization with Zod v4 compatibility, customizable request validation for server adapters, and end-to-end RequestContext support for tracing and datasets/experiments.
Improve tool-call accuracy, MCP fetch hooks receive RequestContext, and reliability updates surface streaming errors, preserve history in stateless tool runs, and clean up orphaned vector embeddings.
Token-aware workspace output, smarter truncation, LSP resolution, sandbox process control, end-to-end auth and RBAC, workflow path tracking, concurrent-safe snapshots, and harness.sendMessage() file support.
A new video course designed to get you from zero to deploying your first agent in under two hours.
Run test cases against agents and workflows, score the results, and track quality over time.
