FrameworksOfficial

AG-UI Mastra 2.0 unifies resumable interrupts and fixes tool-result replay

AG-UI released Mastra adapter 2.0, moving interrupts to the RUN_FINISHED outcome and RunAgentInput.resume while fixing tool-result URLs, cross-turn replay, failed-tool responses, A2UI bridge calls and reasoning attribution.

Why it matters

Continuous generative interfaces depend on reliable pause, approval, resume and tool-result association. Duplicate messages, invalid media encoding or non-resumable threads directly corrupt visible state.

A closer look

The adapter now requires AG-UI 1.0. Image and document URLs remain URL items instead of being encoded as base64, resumed tool calls keep their native message IDs, failed server tools return errors, and A2UI render or bridge-owned calls are answered before the run finishes. It reports at most one interrupt per run and preserves outcomes during late stops.

Who this is for

Developers and platform teams using Mastra, AG-UI, A2UI and frontend tool-approval flows

Original sources

Revisions & corrections (1)

— Read the official AG-UI GitHub release page for release/2026-10-12. Confirmed the 2026-10-12 timestamp, @ag-ui/mastra 2.0.0 package, AG-UI 1.0 requirement, interrupt migration, tool-result URL and replay fixes, A2UI bridge behavior, and listed breaking changes. No conflicting original-source facts were found.

Related events

Frameworks

assistant-ui's October 28 release tightens prompt review, tool-result ownership and configuration writes

assistant-ui released a coordinated package set on October 28. Generative Frame 0.0.9 fills the composer with widget and MCP App prompts for review before sending; assistant-stream 0.3.55 fixes stale model content surviving a later tool response and pending results moving when tool-call IDs are reused; the CLI also narrows configuration paths and plan-revision boundaries.

Frameworks

AI SDK 7.0.155 removes runtime checks already guaranteed by static types

Released October 27, AI SDK 7.0.155 stops revalidating message history against the ModelMessage schema in text and object generation and removes several shape checks covered by TypeScript, reducing a linear cost in long conversations and multi-step runs.