Mastra Scopes settled() to One Thread

// 20m agoPRODUCT UPDATE

Mastra Scopes settled() to One Thread

Mastra’s @mastra/memory adds optional threadId and resourceId scoping to settled(), so callers can await one thread’s Observational Memory work without draining every thread.

// ANALYSIS

Small API change, meaningful concurrency win: `settled()` becomes a precise lifecycle boundary for multi-threaded agent systems.

–Unscoped `settled()` remains global, preserving existing behavior; scoped calls wait only on the specified thread.
–This matters because Observational Memory runs Observer and Reflector tasks in the background, requiring settlement before storage shutdown. [Mastra Memory docs](https://mastra.ai/docs/memory/overview)
–The change aligns with Mastra’s thread-scoped memory direction, while resource-scoped Observational Memory is deprecated. [Observational Memory docs](https://mastra.ai/docs/memory/observational-memory)
–Developers running concurrent agents, test suites, or serverless handlers can now avoid unrelated threads becoming teardown bottlenecks.

// TAGS

mastraagent-memoryframeworkdevtoolopen-source

DISCOVERED

20m ago

2026-11-09

PUBLISHED

1h ago

2026-11-09

RELEVANCE

9/ 10

AUTHOR

mastra

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

INFRA1h ago

Anomaly Teases Private Inference Stack

Anomaly’s OpenCode team previews a private project called Inference, described as “turning machines into tokens,” with a GPU-to-token workflow shown in its README. The post signals an internal inference infrastructure effort, but provides no public code, API, benchmarks, or launch details. [X post](https://x.com/thdxr/status/2119920964607643662)

BENCHMARK1h ago

Claude Fable 5.2 Takes Fifth on MacroscopeBench

Anthropic’s Claude Fable 5.2 scores 78.5 F1 at xhigh on MacroscopeBench, ranking fifth overall and narrowly trailing GPT-6.1 Sol. It also reportedly outperforms Claude Opus 5.5 at matched low, medium, and high effort levels.

INFRA1h ago

OpenAI Control Room adds 35-second kill switch

OpenAI’s Control Room centralizes authorization for model inference across research training, evaluation, and data-processing workloads, allowing people or monitoring systems to halt output within 35 seconds. It fails closed if the control service goes down and is rolling out across OpenAI’s research infrastructure.