Blog — Feb 2026 till Dec 2025
Updates, tutorials, and engineering deep-dives on building AI agents, workflows, and TypeScript tooling with Mastra.
Mastra Changelog 2026-02-11
Observational memory async buffering (default-on) with new structured streaming events, workspace mounts with CompositeFilesystem for unified multi-provider file access.
Announcing Observational Memory
A human-inspired memory system that scores ~95% on LongMemEval — with a completely stable context window.
Introducing Mastra Skills
Mastra Skills teach coding agents about Mastra through structured knowledge files, giving any compatible agent the context it needs to help you build.
Introducing Workspaces: Safe, Controlled Environments for Mastra Agents
File access, command execution, and reusable skills for agents that actually get things done.
Mastra Changelog 2026-02-04
Observational Memory, skills.sh integrations, plus improved tracing and safer error handling.
Human-in-the-Loop: Where to Put Approval in Agents and Workflows
Agents can pause before tool calls. Workflows can suspend at steps. But when they're working together, where's the right place for a human-in-the-loop?
Mastra Changelog 2026-01-30
Unified Workspace API, observability and server adapters improvements.
Human-in-the-Loop: When to Use Agent Approval
When should an agent ask for permission? How to think about approval, suspension, and the trust spectrum in your agent design.
Announcing Mastra 1.0!
APIs are locked, server adapters are the new default, and we've shipped thread cloning, composite storage, AI SDK v6 support, and more.
Mastra Changelog 2026-01-20
Server adapters, composite storage, processor system overhaul, AI SDK v6 support, thread cloning, and more.
Composite Storage: Optimize for Performance, Scale, and Cost with MastraCompositeStore
You can now use multiple storage providers in a single Mastra project, all configured from one place.
Introducing Server Adapters: Run Mastra Inside Your Existing Express or Hono App
Mastra’s new Server Adapters automatically expose your agents, workflows, tools, and MCP servers as HTTP endpoints on the server you’re already running.