@mastra/memory 0.0.0-studio-deploy-20260404184540 → 0.0.0-studio-cli-20260504022012

package/CHANGELOG.md

@@ -1,13 +1,462 @@
 # @mastra/memory
 
-## 0.0.0-studio-deploy-20260404184540
+## 0.0.0-studio-cli-20260504022012
+
+### Patch Changes
+
+- Fixed async reflection buffering incorrectly triggering during idle timeout and provider-change activations when observation tokens are below the reflection activation threshold ([#16076](https://github.com/mastra-ai/mastra/pull/16076))
+
+- Updated dependencies [[`6dcd65f`](https://github.com/mastra-ai/mastra/commit/6dcd65f2a34069e6dc43ba35f1d11119b9b40bef), [`c05c9a1`](https://github.com/mastra-ai/mastra/commit/c05c9a13230988cef6d438a62f37760f31927bc7), [`e24aacb`](https://github.com/mastra-ai/mastra/commit/e24aacba07bd66f5d95b636dc24016fca26b52cf), [`1c2dda8`](https://github.com/mastra-ai/mastra/commit/1c2dda805fbfccc0abf55d4cb20cc34402dc3f0c), [`c721164`](https://github.com/mastra-ai/mastra/commit/c7211643f7ac861f83b19a3757cc921487fc9d75), [`1b55954`](https://github.com/mastra-ai/mastra/commit/1b559541c1e08a10e49d01ffc51a634dfc37a286), [`5adc55e`](https://github.com/mastra-ai/mastra/commit/5adc55e63407be8ee977914957d68bcc2a075ceb), [`70017d7`](https://github.com/mastra-ai/mastra/commit/70017d72ab741b5d7040e2a15c251a317782e39e), [`e4942bc`](https://github.com/mastra-ai/mastra/commit/e4942bc7fdc903572f7d84f26d5e15f9d39c763d)]:
+  - @mastra/core@0.0.0-studio-cli-20260504022012
+
+## 1.17.5-alpha.0
+
+### Patch Changes
+
+- Fixed async reflection buffering incorrectly triggering during idle timeout and provider-change activations when observation tokens are below the reflection activation threshold ([#16076](https://github.com/mastra-ai/mastra/pull/16076))
+
+- Updated dependencies [[`c05c9a1`](https://github.com/mastra-ai/mastra/commit/c05c9a13230988cef6d438a62f37760f31927bc7), [`e24aacb`](https://github.com/mastra-ai/mastra/commit/e24aacba07bd66f5d95b636dc24016fca26b52cf), [`c721164`](https://github.com/mastra-ai/mastra/commit/c7211643f7ac861f83b19a3757cc921487fc9d75), [`1b55954`](https://github.com/mastra-ai/mastra/commit/1b559541c1e08a10e49d01ffc51a634dfc37a286), [`5adc55e`](https://github.com/mastra-ai/mastra/commit/5adc55e63407be8ee977914957d68bcc2a075ceb), [`70017d7`](https://github.com/mastra-ai/mastra/commit/70017d72ab741b5d7040e2a15c251a317782e39e), [`e4942bc`](https://github.com/mastra-ai/mastra/commit/e4942bc7fdc903572f7d84f26d5e15f9d39c763d)]:
+  - @mastra/core@1.32.0-alpha.1
+
+## 1.17.4
+
+### Patch Changes
+
+- Fixed idle timeout and provider-change observation activations blocking on in-progress reflection buffering. These triggers now return immediately, letting the background reflection complete asynchronously. ([#15937](https://github.com/mastra-ai/mastra/pull/15937))
+
+- Updated dependencies [[`920c757`](https://github.com/mastra-ai/mastra/commit/920c75799c6bd71787d86deaf654a35af4c839ca), [`d587199`](https://github.com/mastra-ai/mastra/commit/d5871993c0371bde2b0717d6b47194755baa1443), [`1fe2533`](https://github.com/mastra-ai/mastra/commit/1fe2533c4382ca6858aac7c4b63e888c2eac6541), [`f8694b6`](https://github.com/mastra-ai/mastra/commit/f8694b6fa0b7a5cde71d794c3bbef4957c55bcb8)]:
+  - @mastra/core@1.30.0
+
+## 1.17.4-alpha.0
+
+### Patch Changes
+
+- Fixed idle timeout and provider-change observation activations blocking on in-progress reflection buffering. These triggers now return immediately, letting the background reflection complete asynchronously. ([#15937](https://github.com/mastra-ai/mastra/pull/15937))
+
+- Updated dependencies [[`d587199`](https://github.com/mastra-ai/mastra/commit/d5871993c0371bde2b0717d6b47194755baa1443)]:
+  - @mastra/core@1.29.2-alpha.0
+
+## 1.17.3
+
+### Patch Changes
+
+- Fixed memory package startup with older core versions. ([#15772](https://github.com/mastra-ai/mastra/pull/15772))
+
+- Updated dependencies [[`6db978c`](https://github.com/mastra-ai/mastra/commit/6db978c42e94e75540a504f7230086f0b5cd35f9), [`512a013`](https://github.com/mastra-ai/mastra/commit/512a013f285aa9c0aa8f08a35b2ce09f9938b017), [`e9becde`](https://github.com/mastra-ai/mastra/commit/e9becdeed9176b9f8392e557bde12b933f99cf7a), [`703a443`](https://github.com/mastra-ai/mastra/commit/703a44390c587d9c0b8ae94ec4edd8afb2a74044), [`808df1b`](https://github.com/mastra-ai/mastra/commit/808df1b39358b5f10b7317107e42b1fda7c87185)]:
+  - @mastra/core@1.29.1
+
+## 1.17.3-alpha.0
+
+### Patch Changes
+
+- Fixed memory package startup with older core versions. ([#15772](https://github.com/mastra-ai/mastra/pull/15772))
+
+## 1.17.2
+
+### Patch Changes
+
+- Fixed the recall tool so message browsing uses the current thread by default and explains when it does. ([#15807](https://github.com/mastra-ai/mastra/pull/15807))
+
+- Updated dependencies [[`28caa5b`](https://github.com/mastra-ai/mastra/commit/28caa5b032358545af2589ed90636eccb4dd9d2f), [`c1ae974`](https://github.com/mastra-ai/mastra/commit/c1ae97491f6e57378ce880c3a397778c42adcdf1), [`b510d36`](https://github.com/mastra-ai/mastra/commit/b510d368f73dab6be2e2c2bc99035aaef1fb7d7a), [`13b4d7c`](https://github.com/mastra-ai/mastra/commit/13b4d7c16de34dff9095d1cd80f22f544b6cfe75), [`7a7b313`](https://github.com/mastra-ai/mastra/commit/7a7b3138fb3bcf0b0c740eaea07971e43d330ef3), [`c04417b`](https://github.com/mastra-ai/mastra/commit/c04417ba0a2e4ded66da4352331ef29cd4bd1d79), [`cf25a03`](https://github.com/mastra-ai/mastra/commit/cf25a03132164b9dc1e5dccf7394824e33007c51), [`8a71261`](https://github.com/mastra-ai/mastra/commit/8a71261e3954ae617c6f8e25767b951f99438ab2), [`9e973b0`](https://github.com/mastra-ai/mastra/commit/9e973b010dacfa15ac82b0072897319f5234b90a), [`dd934a0`](https://github.com/mastra-ai/mastra/commit/dd934a0982ce0f78712fbd559e4f2410bf594b39), [`ba6b0c5`](https://github.com/mastra-ai/mastra/commit/ba6b0c51bfce358554fd33c7f2bcd5593633f2ff), [`a6dac0a`](https://github.com/mastra-ai/mastra/commit/a6dac0a40c7181161b1add4e8534f962bcbc9aa7), [`5a4b1ee`](https://github.com/mastra-ai/mastra/commit/5a4b1ee80212969621228104995589c0fa59e575), [`5a4b1ee`](https://github.com/mastra-ai/mastra/commit/5a4b1ee80212969621228104995589c0fa59e575), [`5a4b1ee`](https://github.com/mastra-ai/mastra/commit/5a4b1ee80212969621228104995589c0fa59e575), [`6c8c6c7`](https://github.com/mastra-ai/mastra/commit/6c8c6c71518394321a4692614aa4b11f3bb0a343), [`5a4b1ee`](https://github.com/mastra-ai/mastra/commit/5a4b1ee80212969621228104995589c0fa59e575), [`7d056b6`](https://github.com/mastra-ai/mastra/commit/7d056b6ecf603cacaa0f663ff1df025ed885b6c1), [`9cef83b`](https://github.com/mastra-ai/mastra/commit/9cef83b8a642b8098747772921e3523b492bafbc), [`d30e215`](https://github.com/mastra-ai/mastra/commit/d30e2156c746bc9fd791745cec1cc24377b66789), [`021a60f`](https://github.com/mastra-ai/mastra/commit/021a60f1f3e0135a70ef23c58be7a9b3aaffe6b4), [`73f2809`](https://github.com/mastra-ai/mastra/commit/73f2809721db24e98cdf122539652a455211b450), [`aedeea4`](https://github.com/mastra-ai/mastra/commit/aedeea48a94f728323f040478775076b9574be50), [`26f1f94`](https://github.com/mastra-ai/mastra/commit/26f1f9490574b864ba1ecedf2c9632e0767a23bd), [`8126d86`](https://github.com/mastra-ai/mastra/commit/8126d8638411eacfafdc29036ac998e8757ea66f), [`73b45fa`](https://github.com/mastra-ai/mastra/commit/73b45facdef4fbcb8af710c50f0646f18619dbaa), [`ae97520`](https://github.com/mastra-ai/mastra/commit/ae975206fdb0f6ef03c4d5bf94f7dc7c3f706c02), [`7a7b313`](https://github.com/mastra-ai/mastra/commit/7a7b3138fb3bcf0b0c740eaea07971e43d330ef3), [`441670a`](https://github.com/mastra-ai/mastra/commit/441670a02c9dc7731c52674f55481e7848a84523)]:
+  - @mastra/core@1.29.0
+
+## 1.17.2-alpha.0
+
+### Patch Changes
+
+- Fixed Observational Memory duplicate assistant messages when a thread is rerun or an agent continues a long-running loop. ([#15774](https://github.com/mastra-ai/mastra/pull/15774))
+
+- Fixed the recall tool so message browsing uses the current thread by default and explains when it does. ([#15807](https://github.com/mastra-ai/mastra/pull/15807))
+
+- Updated dependencies [[`c1ae974`](https://github.com/mastra-ai/mastra/commit/c1ae97491f6e57378ce880c3a397778c42adcdf1), [`13b4d7c`](https://github.com/mastra-ai/mastra/commit/13b4d7c16de34dff9095d1cd80f22f544b6cfe75), [`5a4b1ee`](https://github.com/mastra-ai/mastra/commit/5a4b1ee80212969621228104995589c0fa59e575), [`5a4b1ee`](https://github.com/mastra-ai/mastra/commit/5a4b1ee80212969621228104995589c0fa59e575), [`5a4b1ee`](https://github.com/mastra-ai/mastra/commit/5a4b1ee80212969621228104995589c0fa59e575), [`6c8c6c7`](https://github.com/mastra-ai/mastra/commit/6c8c6c71518394321a4692614aa4b11f3bb0a343), [`5a4b1ee`](https://github.com/mastra-ai/mastra/commit/5a4b1ee80212969621228104995589c0fa59e575), [`ec4cb26`](https://github.com/mastra-ai/mastra/commit/ec4cb26919972eb2031fea510f8f013e1d5b7ee2)]:
+  - @mastra/core@1.29.0-alpha.6
+
+## 1.17.1
+
+### Patch Changes
+
+- Fixed streamed tool results being replayed when observational memory runs mid-stream. ([#15701](https://github.com/mastra-ai/mastra/pull/15701))
+  Fixed observational memory markers being saved as separate empty assistant messages.
+- Updated dependencies [[`733bf53`](https://github.com/mastra-ai/mastra/commit/733bf53d9352aedd3ef38c3d501edb275b65b43c), [`5405b3b`](https://github.com/mastra-ai/mastra/commit/5405b3b35325c5b8fb34fc7ac109bd2feb7bb6fe), [`45e29cb`](https://github.com/mastra-ai/mastra/commit/45e29cb5b5737f3083eb3852db02b944b9cf37ed), [`750b4d3`](https://github.com/mastra-ai/mastra/commit/750b4d3d8231f92e769b2c485921ac5a8ca639b9), [`c321127`](https://github.com/mastra-ai/mastra/commit/c3211275fc195de9ad1ead2746b354beb8eae6e8), [`a07bcef`](https://github.com/mastra-ai/mastra/commit/a07bcefea77c03d6d322caad973dca49b4b15fa1), [`696694e`](https://github.com/mastra-ai/mastra/commit/696694e00f29241a25dd1a1b749afa06c3a626b4), [`b084a80`](https://github.com/mastra-ai/mastra/commit/b084a800db0f82d62e1fc3d6e3e3480da1ba5a53), [`82b7a96`](https://github.com/mastra-ai/mastra/commit/82b7a964169636c1d1e0c694fc892a213b0179d5), [`df97812`](https://github.com/mastra-ai/mastra/commit/df97812bd949dcafeb074b80ecab501724b49c3b), [`8bbe360`](https://github.com/mastra-ai/mastra/commit/8bbe36042af7fc4be0244dffd8913f6795179421), [`f6b8ba8`](https://github.com/mastra-ai/mastra/commit/f6b8ba8dbf533b7a8db90c72b6805ddc804a3a72), [`a07bcef`](https://github.com/mastra-ai/mastra/commit/a07bcefea77c03d6d322caad973dca49b4b15fa1)]:
+  - @mastra/core@1.28.0
+
+## 1.17.1-alpha.0
+
+### Patch Changes
+
+- Fixed observational memory message boundaries so streamed tool results continue in a fresh assistant message instead of being filtered out mid-run, and OM lifecycle markers persist through memory without creating data-only database messages. ([#15701](https://github.com/mastra-ai/mastra/pull/15701))
+
+- Updated dependencies [[`750b4d3`](https://github.com/mastra-ai/mastra/commit/750b4d3d8231f92e769b2c485921ac5a8ca639b9)]:
+  - @mastra/core@1.28.0-alpha.1
+
+## 1.17.0
+
+### Minor Changes
+
+- Added opt-in temporal-gap markers for observational memory. When enabled via `observationalMemory.temporalMarkers: true`, the agent receives a `<system-reminder type="temporal-gap">` before any user message that arrives more than 10 minutes after the previous one, so it can anchor responses in real elapsed time. Markers are persisted, surfaced to the observer, and rendered by the MastraCode TUI on reload. ([#15605](https://github.com/mastra-ai/mastra/pull/15605))
+
+### Patch Changes
+
+- Fixed observer agent truncation that could cut UTF-16 surrogate pairs in half when formatting messages, tool results, or observation lines with emoji or other astral-plane characters. This produced lone surrogates that strict JSON parsers (including Anthropic's) reject with errors like `no low surrogate in string`, causing observer runs to fail. ([#15634](https://github.com/mastra-ai/mastra/pull/15634))
+
+- Updated dependencies [[`f112db1`](https://github.com/mastra-ai/mastra/commit/f112db179557ae9b5a0f1d25dc47f928d7d61cd9), [`21d9706`](https://github.com/mastra-ai/mastra/commit/21d970604d89eee970cbf8013d26d7551aff6ea5), [`0a0aa94`](https://github.com/mastra-ai/mastra/commit/0a0aa94729592e99885af2efb90c56aaada62247), [`ed07df3`](https://github.com/mastra-ai/mastra/commit/ed07df32a9d539c8261e892fc1bade783f5b41a6), [`01a7d51`](https://github.com/mastra-ai/mastra/commit/01a7d513493d21562f677f98550f7ceb165ba78c)]:
+  - @mastra/core@1.27.0
+
+## 1.17.0-alpha.0
+
+### Minor Changes
+
+- Added opt-in temporal-gap markers for observational memory. When enabled via `observationalMemory.temporalMarkers: true`, the agent receives a `<system-reminder type="temporal-gap">` before any user message that arrives more than 10 minutes after the previous one, so it can anchor responses in real elapsed time. Markers are persisted, surfaced to the observer, and rendered by the MastraCode TUI on reload. ([#15605](https://github.com/mastra-ai/mastra/pull/15605))
+
+### Patch Changes
+
+- Fixed observer agent truncation that could cut UTF-16 surrogate pairs in half when formatting messages, tool results, or observation lines with emoji or other astral-plane characters. This produced lone surrogates that strict JSON parsers (including Anthropic's) reject with errors like `no low surrogate in string`, causing observer runs to fail. ([#15634](https://github.com/mastra-ai/mastra/pull/15634))
+
+- Updated dependencies [[`0a0aa94`](https://github.com/mastra-ai/mastra/commit/0a0aa94729592e99885af2efb90c56aaada62247), [`01a7d51`](https://github.com/mastra-ai/mastra/commit/01a7d513493d21562f677f98550f7ceb165ba78c)]:
+  - @mastra/core@1.27.0-alpha.1
+
+## 1.16.0
+
+### Minor Changes
+
+- Added activateAfterIdle setting for observational memory so buffered observations can activate after idle time before the next prompt. ([#15365](https://github.com/mastra-ai/mastra/pull/15365))
+
+  **Example**
+
+  Set `activateAfterIdle: 300_000` (or `"5m"`) on the `observationalMemory` config to activate buffered context after 5 minutes of inactivity.
+
+  This helps long-running threads reuse compressed context after prompt cache TTLs expire instead of sending a larger raw message window on the next request.
+
+- Added `activateOnProviderChange` so observational memory can activate buffered observations and reflections before switching to a different provider or model. ([#15420](https://github.com/mastra-ai/mastra/pull/15420))
+
+  ```ts
+  const memory = new Memory({
+    options: {
+      observationalMemory: {
+        model: 'google/gemini-2.5-flash',
+        activateOnProviderChange: true,
+      },
+    },
+  });
+  ```
+
+  This helps keep prompt-cache savings when the next step cannot reuse the previous provider's cache.
+
+### Patch Changes
+
+- Fixed early observational memory activations so buffered reflections are only activated when they still leave a healthy active observation set. ([#15462](https://github.com/mastra-ai/mastra/pull/15462))
+
+  Before this change, idle-timeout (`activateAfterIdle`) and model/provider-change (`activateOnProviderChange`) activations could swap in a buffered reflection too early. In bad cases, that replaced a large raw observation tail with a much smaller mostly-compressed result, which hurt reflection quality.
+
+  Early activations now stay buffered unless both of these checks pass:
+  - The unreflected observation tail is at least as large as the buffered reflection, so the activated result is not dominated by compressed content.
+  - The combined post-activation size is at least 75% of what a normal threshold activation would produce, so early activations do not cliff far below the regular target.
+
+  This update also fixes false `provider_change` activations when older persisted messages only contain a bare model id like `gpt-5.4` while newer turns use the fully qualified `provider/modelId` form.
+
+- Fixed a security issue where several parsing and tracing paths could slow down on malformed or attacker-crafted input. Normal behavior is unchanged, and these packages now handle pathological input in linear time. ([#15566](https://github.com/mastra-ai/mastra/pull/15566))
+
+- Fixed other-thread context filtering falling back to the observational memory record timestamp when thread metadata is missing. ([#15269](https://github.com/mastra-ai/mastra/pull/15269))
+
+- Updated dependencies [[`20f59b8`](https://github.com/mastra-ai/mastra/commit/20f59b876cf91199efbc49a0e36b391240708f08), [`aba393e`](https://github.com/mastra-ai/mastra/commit/aba393e2da7390c69b80e516a4f153cda6f09376), [`3d83d06`](https://github.com/mastra-ai/mastra/commit/3d83d06f776f00fb5f4163dddd32a030c5c20844), [`e2687a7`](https://github.com/mastra-ai/mastra/commit/e2687a7408790c384563816a9a28ed06735684c9), [`fdd54cf`](https://github.com/mastra-ai/mastra/commit/fdd54cf612a9af876e9fdd85e534454f6e7dd518), [`6315317`](https://github.com/mastra-ai/mastra/commit/63153175fe9a7b224e5be7c209bbebc01dd9b0d5), [`a371ac5`](https://github.com/mastra-ai/mastra/commit/a371ac534aa1bb368a1acf9d8b313378dfdc787e), [`7db42a9`](https://github.com/mastra-ai/mastra/commit/7db42a9cccd3b29c44fb0731f792c51575e8421c), [`0474c2b`](https://github.com/mastra-ai/mastra/commit/0474c2b2e7c7e1ad8691dca031284841391ff1ef), [`0a5fa1d`](https://github.com/mastra-ai/mastra/commit/0a5fa1d3cb0583889d06687155f26fd7d2edc76c), [`7e0e63e`](https://github.com/mastra-ai/mastra/commit/7e0e63e2e485e84442351f4c7a79a424c83539dc), [`ea43e64`](https://github.com/mastra-ai/mastra/commit/ea43e646dd95d507694b6112b0bf1df22ad552b2), [`f607106`](https://github.com/mastra-ai/mastra/commit/f607106854c6416c4a07d4082604b9f66d047221), [`30456b6`](https://github.com/mastra-ai/mastra/commit/30456b6b08c8fd17e109dd093b73d93b65e83bc5), [`9d11a8c`](https://github.com/mastra-ai/mastra/commit/9d11a8c1c8924eb975a245a5884d40ca1b7e0491), [`9d3b24b`](https://github.com/mastra-ai/mastra/commit/9d3b24b19407ae9c09586cf7766d38dc4dff4a69), [`00d1b16`](https://github.com/mastra-ai/mastra/commit/00d1b16b401199cb294fa23f43336547db4dca9b), [`47cee3e`](https://github.com/mastra-ai/mastra/commit/47cee3e137fe39109cf7fffd2a8cf47b76dc702e), [`62919a6`](https://github.com/mastra-ai/mastra/commit/62919a6ee0fbf3779ad21a97b1ec6696515d5104), [`d246696`](https://github.com/mastra-ai/mastra/commit/d246696139a3144a5b21b042d41c532688e957e1), [`354f9ce`](https://github.com/mastra-ai/mastra/commit/354f9ce1ca6af2074b6a196a23f8ec30012dccca), [`16e34ca`](https://github.com/mastra-ai/mastra/commit/16e34caa98b9a114b17a6125e4e3fd87f169d0d0), [`7020c06`](https://github.com/mastra-ai/mastra/commit/7020c0690b199d9da337f0e805f16948e557922e), [`8786a61`](https://github.com/mastra-ai/mastra/commit/8786a61fa54ba265f85eeff9985ca39863d18bb6), [`9467ea8`](https://github.com/mastra-ai/mastra/commit/9467ea87695749a53dfc041576410ebf9ee7bb67), [`7338d94`](https://github.com/mastra-ai/mastra/commit/7338d949380cf68b095342e8e42610dc51d557c1), [`c80dc16`](https://github.com/mastra-ai/mastra/commit/c80dc16e113e6cc159f510ffde501ad4711b2189), [`af8a57e`](https://github.com/mastra-ai/mastra/commit/af8a57ed9ba9685ad8601d5b71ae3706da6222f9), [`d63ffdb`](https://github.com/mastra-ai/mastra/commit/d63ffdbb2c11e76fe5ea45faab44bc15460f010c), [`47cee3e`](https://github.com/mastra-ai/mastra/commit/47cee3e137fe39109cf7fffd2a8cf47b76dc702e), [`1bd5104`](https://github.com/mastra-ai/mastra/commit/1bd51048b6da93507276d6623e3fd96a9e1a8944), [`e9837b5`](https://github.com/mastra-ai/mastra/commit/e9837b53699e18711b09e0ca010a4106376f2653), [`c65aec3`](https://github.com/mastra-ai/mastra/commit/c65aec356cc037ee7c4b30ccea946807d4c4f443), [`8f1b280`](https://github.com/mastra-ai/mastra/commit/8f1b280b7fe6999ec654f160cb69c1a8719e7a57), [`92dcf02`](https://github.com/mastra-ai/mastra/commit/92dcf029294210ac91b090900c1a0555a425c57a), [`0fd90a2`](https://github.com/mastra-ai/mastra/commit/0fd90a215caf5fca8099c15a67ca03e4427747a3), [`8fb2405`](https://github.com/mastra-ai/mastra/commit/8fb2405138f2d208b7962ad03f121ca25bcc28c5), [`12df98c`](https://github.com/mastra-ai/mastra/commit/12df98c4904643d9481f5c78f3bed443725b4c96)]:
+  - @mastra/core@1.26.0
+  - @mastra/schema-compat@1.2.9
+
+## 1.16.0-alpha.4
+
+### Patch Changes
+
+- Fixed other-thread context filtering falling back to the observational memory record timestamp when thread metadata is missing. ([#15269](https://github.com/mastra-ai/mastra/pull/15269))
+
+- Updated dependencies [[`a371ac5`](https://github.com/mastra-ai/mastra/commit/a371ac534aa1bb368a1acf9d8b313378dfdc787e), [`47cee3e`](https://github.com/mastra-ai/mastra/commit/47cee3e137fe39109cf7fffd2a8cf47b76dc702e), [`c80dc16`](https://github.com/mastra-ai/mastra/commit/c80dc16e113e6cc159f510ffde501ad4711b2189), [`47cee3e`](https://github.com/mastra-ai/mastra/commit/47cee3e137fe39109cf7fffd2a8cf47b76dc702e)]:
+  - @mastra/core@1.26.0-alpha.12
+
+## 1.16.0-alpha.3
+
+### Patch Changes
+
+- Fixed a security issue where several parsing and tracing paths could slow down on malformed or attacker-crafted input. Normal behavior is unchanged, and these packages now handle pathological input in linear time. ([#15566](https://github.com/mastra-ai/mastra/pull/15566))
+
+- Updated dependencies [[`aba393e`](https://github.com/mastra-ai/mastra/commit/aba393e2da7390c69b80e516a4f153cda6f09376), [`0a5fa1d`](https://github.com/mastra-ai/mastra/commit/0a5fa1d3cb0583889d06687155f26fd7d2edc76c), [`ea43e64`](https://github.com/mastra-ai/mastra/commit/ea43e646dd95d507694b6112b0bf1df22ad552b2), [`00d1b16`](https://github.com/mastra-ai/mastra/commit/00d1b16b401199cb294fa23f43336547db4dca9b), [`af8a57e`](https://github.com/mastra-ai/mastra/commit/af8a57ed9ba9685ad8601d5b71ae3706da6222f9)]:
+  - @mastra/core@1.26.0-alpha.10
+
+## 1.16.0-alpha.2
+
+### Minor Changes
+
+- Added `activateOnProviderChange` so observational memory can activate buffered observations and reflections before switching to a different provider or model. ([#15420](https://github.com/mastra-ai/mastra/pull/15420))
+
+  ```ts
+  const memory = new Memory({
+    options: {
+      observationalMemory: {
+        model: 'google/gemini-2.5-flash',
+        activateOnProviderChange: true,
+      },
+    },
+  });
+  ```
+
+  This helps keep prompt-cache savings when the next step cannot reuse the previous provider's cache.
+
+### Patch Changes
+
+- Fixed early observational memory activations so buffered reflections are only activated when they still leave a healthy active observation set. ([#15462](https://github.com/mastra-ai/mastra/pull/15462))
+
+  Before this change, idle-timeout (`activateAfterIdle`) and model/provider-change (`activateOnProviderChange`) activations could swap in a buffered reflection too early. In bad cases, that replaced a large raw observation tail with a much smaller mostly-compressed result, which hurt reflection quality.
+
+  Early activations now stay buffered unless both of these checks pass:
+  - The unreflected observation tail is at least as large as the buffered reflection, so the activated result is not dominated by compressed content.
+  - The combined post-activation size is at least 75% of what a normal threshold activation would produce, so early activations do not cliff far below the regular target.
+
+  This update also fixes false `provider_change` activations when older persisted messages only contain a bare model id like `gpt-5.4` while newer turns use the fully qualified `provider/modelId` form.
+
+- Updated dependencies [[`0474c2b`](https://github.com/mastra-ai/mastra/commit/0474c2b2e7c7e1ad8691dca031284841391ff1ef), [`f607106`](https://github.com/mastra-ai/mastra/commit/f607106854c6416c4a07d4082604b9f66d047221), [`62919a6`](https://github.com/mastra-ai/mastra/commit/62919a6ee0fbf3779ad21a97b1ec6696515d5104), [`0fd90a2`](https://github.com/mastra-ai/mastra/commit/0fd90a215caf5fca8099c15a67ca03e4427747a3)]:
+  - @mastra/core@1.26.0-alpha.4
+
+## 1.16.0-alpha.1
+
+### Patch Changes
+
+- Updated dependencies [[`fdd54cf`](https://github.com/mastra-ai/mastra/commit/fdd54cf612a9af876e9fdd85e534454f6e7dd518), [`7db42a9`](https://github.com/mastra-ai/mastra/commit/7db42a9cccd3b29c44fb0731f792c51575e8421c), [`30456b6`](https://github.com/mastra-ai/mastra/commit/30456b6b08c8fd17e109dd093b73d93b65e83bc5), [`9d11a8c`](https://github.com/mastra-ai/mastra/commit/9d11a8c1c8924eb975a245a5884d40ca1b7e0491), [`d246696`](https://github.com/mastra-ai/mastra/commit/d246696139a3144a5b21b042d41c532688e957e1), [`354f9ce`](https://github.com/mastra-ai/mastra/commit/354f9ce1ca6af2074b6a196a23f8ec30012dccca), [`e9837b5`](https://github.com/mastra-ai/mastra/commit/e9837b53699e18711b09e0ca010a4106376f2653)]:
+  - @mastra/core@1.26.0-alpha.3
+  - @mastra/schema-compat@1.2.9-alpha.1
+
+## 1.16.0-alpha.0
+
+### Minor Changes
+
+- Added activateAfterIdle setting for observational memory so buffered observations can activate after idle time before the next prompt. ([#15365](https://github.com/mastra-ai/mastra/pull/15365))
+
+  **Example**
+
+  Set `activateAfterIdle: 300_000` (or `"5m"`) on the `observationalMemory` config to activate buffered context after 5 minutes of inactivity.
+
+  This helps long-running threads reuse compressed context after prompt cache TTLs expire instead of sending a larger raw message window on the next request.
+
+### Patch Changes
+
+- Updated dependencies [[`3d83d06`](https://github.com/mastra-ai/mastra/commit/3d83d06f776f00fb5f4163dddd32a030c5c20844), [`7e0e63e`](https://github.com/mastra-ai/mastra/commit/7e0e63e2e485e84442351f4c7a79a424c83539dc), [`9467ea8`](https://github.com/mastra-ai/mastra/commit/9467ea87695749a53dfc041576410ebf9ee7bb67), [`7338d94`](https://github.com/mastra-ai/mastra/commit/7338d949380cf68b095342e8e42610dc51d557c1), [`c65aec3`](https://github.com/mastra-ai/mastra/commit/c65aec356cc037ee7c4b30ccea946807d4c4f443)]:
+  - @mastra/core@1.26.0-alpha.2
+  - @mastra/schema-compat@1.2.9-alpha.0
+
+## 1.15.1
+
+### Patch Changes
+
+- Fixed gateway model detection to use duck typing instead of instanceof check, preventing potential failures from cross-package module resolution issues. Propagates `gatewayId` through the AISDKV5LanguageModel wrapper so duck-type detection works even when models are re-wrapped. ([#15168](https://github.com/mastra-ai/mastra/pull/15168))
+
+- Updated dependencies [[`87df955`](https://github.com/mastra-ai/mastra/commit/87df955c028660c075873fd5d74af28233ce32eb), [`8fad147`](https://github.com/mastra-ai/mastra/commit/8fad14759804179c8e080ce4d9dec6ef1a808b31), [`582644c`](https://github.com/mastra-ai/mastra/commit/582644c4a87f83b4f245a84d72b9e8590585012e), [`cbdf3e1`](https://github.com/mastra-ai/mastra/commit/cbdf3e12b3d0c30a6e5347be658e2009648c130a), [`8fe46d3`](https://github.com/mastra-ai/mastra/commit/8fe46d354027f3f0f0846e64219772348de106dd), [`18c67db`](https://github.com/mastra-ai/mastra/commit/18c67dbb9c9ebc26f26f65f7d3ff836e5691ef46), [`4ba3bb1`](https://github.com/mastra-ai/mastra/commit/4ba3bb1e465ad2ddaba3bbf2bc47e0faec32985e), [`5d84914`](https://github.com/mastra-ai/mastra/commit/5d84914e0e520c642a40329b210b413fcd139898), [`8dcc77e`](https://github.com/mastra-ai/mastra/commit/8dcc77e78a5340f5848f74b9e9f1b3da3513c1f5), [`aa67fc5`](https://github.com/mastra-ai/mastra/commit/aa67fc59ee8a5eeff1f23eb05970b8d7a536c8ff), [`fd2f314`](https://github.com/mastra-ai/mastra/commit/fd2f31473d3449b6b97e837ef8641264377f41a7), [`fa8140b`](https://github.com/mastra-ai/mastra/commit/fa8140bcd4251d2e3ac85fdc5547dfc4f372b5be), [`190f452`](https://github.com/mastra-ai/mastra/commit/190f45258b0640e2adfc8219fa3258cdc5b8f071), [`e80fead`](https://github.com/mastra-ai/mastra/commit/e80fead1412cc0d1b2f7d6a1ce5017d9e0098ff7), [`0287b64`](https://github.com/mastra-ai/mastra/commit/0287b644a5c3272755cf3112e71338106664103b), [`7e7bf60`](https://github.com/mastra-ai/mastra/commit/7e7bf606886bf374a6f9d4ca9b09dd83d0533372), [`184907d`](https://github.com/mastra-ai/mastra/commit/184907d775d8609c03c26e78ccaf37315f3aa287), [`075e91a`](https://github.com/mastra-ai/mastra/commit/075e91a4549baf46ad7a42a6a8ac8dfa78cc09e6), [`2a69802`](https://github.com/mastra-ai/mastra/commit/2a69802a0fc6d8a25a77fa6a42276e9d59a83914), [`0c4cd13`](https://github.com/mastra-ai/mastra/commit/0c4cd131931c04ac5405373c932a242dbe88edd6), [`b16a753`](https://github.com/mastra-ai/mastra/commit/b16a753d5748440248d7df82e29bb987a9c8386c)]:
+  - @mastra/core@1.25.0
+  - @mastra/schema-compat@1.2.8
+
+## 1.15.1-alpha.1
+
+### Patch Changes
+
+- Updated dependencies [[`4ba3bb1`](https://github.com/mastra-ai/mastra/commit/4ba3bb1e465ad2ddaba3bbf2bc47e0faec32985e), [`2a69802`](https://github.com/mastra-ai/mastra/commit/2a69802a0fc6d8a25a77fa6a42276e9d59a83914)]:
+  - @mastra/core@1.25.0-alpha.2
+  - @mastra/schema-compat@1.2.8-alpha.0
+
+## 1.15.1-alpha.0
+
+### Patch Changes
+
+- Fixed gateway model detection to use duck typing instead of instanceof check, preventing potential failures from cross-package module resolution issues. Propagates `gatewayId` through the AISDKV5LanguageModel wrapper so duck-type detection works even when models are re-wrapped. ([#15168](https://github.com/mastra-ai/mastra/pull/15168))
+
+- Updated dependencies [[`87df955`](https://github.com/mastra-ai/mastra/commit/87df955c028660c075873fd5d74af28233ce32eb), [`075e91a`](https://github.com/mastra-ai/mastra/commit/075e91a4549baf46ad7a42a6a8ac8dfa78cc09e6)]:
+  - @mastra/core@1.24.2-alpha.0
+
+## 1.15.0
+
+### Minor Changes
+
+- Updated the recall tool to support more precise message browsing for agents. ([#15116](https://github.com/mastra-ai/mastra/pull/15116))
+
+  Agents using `recall` can now pass `partType` and `toolName` to narrow message results to specific parts, such as tool calls or tool results for one tool. This change also adds `threadId: "current"` support across recall modes and `anchor: "start" | "end"` for no-cursor message paging, making it easier to inspect recent thread activity and past tool usage.
+
+### Patch Changes
+
+- Fixed reflection threshold not respecting per-record overrides set via the PATCH API. Previously, lowering the reflection threshold for a specific record had no effect on the actual reflection trigger — only the default 40k threshold was used. Now per-record overrides are correctly applied in both sync and async reflection paths. ([#15170](https://github.com/mastra-ai/mastra/pull/15170))
+
+- Improved observational memory formatting to use part timestamps when rendering dates and times. ([#15121](https://github.com/mastra-ai/mastra/pull/15121))
+
+  Observer history now follows part-level timing more closely, so the rendered memory context is more accurate when messages contain parts created at different times.
+
+- Fixed message history doubling when using Observational Memory with the Mastra gateway. The local ObservationalMemoryProcessor now detects when the agent's model is routed through the Mastra gateway and skips its input/output processing, since the gateway handles OM server-side. ([#15161](https://github.com/mastra-ai/mastra/pull/15161))
+
+- Updated dependencies [[`8db7663`](https://github.com/mastra-ai/mastra/commit/8db7663c9a9c735828094c359d2e327fd4f8fba3), [`153e864`](https://github.com/mastra-ai/mastra/commit/153e86476b425db7cd0dc8490050096e92964a38), [`715710d`](https://github.com/mastra-ai/mastra/commit/715710d12fa47cf88e09d41f13843eddc29327b0), [`378c6c4`](https://github.com/mastra-ai/mastra/commit/378c6c4755726e8d8cf83a14809b350b90d46c62), [`9f91fd5`](https://github.com/mastra-ai/mastra/commit/9f91fd538ab2a44f8cc740bcad8e51205f74fbea), [`ba6fa9c`](https://github.com/mastra-ai/mastra/commit/ba6fa9cc0f3e1912c49fd70d4c3bb8c44903ddaa)]:
+  - @mastra/core@1.24.0
+
+## 1.15.0-alpha.3
+
+### Patch Changes
+
+- Fixed reflection threshold not respecting per-record overrides set via the PATCH API. Previously, lowering the reflection threshold for a specific record had no effect on the actual reflection trigger — only the default 40k threshold was used. Now per-record overrides are correctly applied in both sync and async reflection paths. ([#15170](https://github.com/mastra-ai/mastra/pull/15170))
+
+## 1.15.0-alpha.2
+
+### Patch Changes
+
+- Fixed message history doubling when using Observational Memory with the Mastra gateway. The local ObservationalMemoryProcessor now detects when the agent's model is routed through the Mastra gateway and skips its input/output processing, since the gateway handles OM server-side. ([#15161](https://github.com/mastra-ai/mastra/pull/15161))
+
+## 1.15.0-alpha.1
+
+### Patch Changes
+
+- Improved observational memory formatting to use part timestamps when rendering dates and times. ([#15121](https://github.com/mastra-ai/mastra/pull/15121))
+
+  Observer history now follows part-level timing more closely, so the rendered memory context is more accurate when messages contain parts created at different times.
+
+- Updated dependencies [[`8db7663`](https://github.com/mastra-ai/mastra/commit/8db7663c9a9c735828094c359d2e327fd4f8fba3), [`715710d`](https://github.com/mastra-ai/mastra/commit/715710d12fa47cf88e09d41f13843eddc29327b0), [`378c6c4`](https://github.com/mastra-ai/mastra/commit/378c6c4755726e8d8cf83a14809b350b90d46c62), [`9f91fd5`](https://github.com/mastra-ai/mastra/commit/9f91fd538ab2a44f8cc740bcad8e51205f74fbea), [`ba6fa9c`](https://github.com/mastra-ai/mastra/commit/ba6fa9cc0f3e1912c49fd70d4c3bb8c44903ddaa)]:
+  - @mastra/core@1.24.0-alpha.1
+
+## 1.15.0-alpha.0
+
+### Minor Changes
+
+- Updated the recall tool to support more precise message browsing for agents. ([#15116](https://github.com/mastra-ai/mastra/pull/15116))
+
+  Agents using `recall` can now pass `partType` and `toolName` to narrow message results to specific parts, such as tool calls or tool results for one tool. This change also adds `threadId: "current"` support across recall modes and `anchor: "start" | "end"` for no-cursor message paging, making it easier to inspect recent thread activity and past tool usage.
+
+## 1.14.0
+
+### Minor Changes
+
+- Added usage data to ObserveHooks callbacks and standalone reflect() return. ([#15047](https://github.com/mastra-ai/mastra/pull/15047))
+
+  **ObserveHooks:** `onObservationEnd` and `onReflectionEnd` now receive a result object containing token usage from the underlying LLM call. This enables reliable usage tracking across all observation and reflection paths (sync, async buffered, and resource-scoped).
+
+  **Standalone reflect():** `reflect()` now returns `{ reflected, record, usage? }` so callers can capture token usage without hooks.
+
+  **Examples**
+
+  ```ts
+  // Via hooks
+  await memory.observe({
+    threadId,
+    messages,
+    hooks: {
+      onObservationEnd: ({ usage }) => {
+        // usage: { inputTokens, outputTokens, totalTokens }
+      },
+      onReflectionEnd: ({ usage }) => {
+        // usage: { inputTokens, outputTokens, totalTokens }
+      },
+    },
+  });
+
+  // Via standalone reflect()
+  const { reflected, usage } = await memory.reflect(threadId, resourceId);
+  ```
+
+  Existing callbacks that accept no arguments continue to work without changes.
+
+- Added tracing support to Memory operations (recall, save, delete, update working memory). When an `observabilityContext` is provided, Memory creates `MEMORY_OPERATION` spans that capture operation type, message counts, embedding token usage, and vector result counts. Tracing is fully opt-in — existing usage without `observabilityContext` is unaffected. ([#14305](https://github.com/mastra-ai/mastra/pull/14305))
+
+  **Example usage:**
+
+  ```typescript
+  import { Memory } from '@mastra/memory';
+  import { InMemoryStore } from '@mastra/core/storage';
+
+  const memory = new Memory({ storage: new InMemoryStore() });
+
+  // Pass observabilityContext to create observable spans
+  await memory.recall({
+    threadId: 'thread-1',
+    observabilityContext: { tracingContext: { currentSpan: parentSpan } },
+  });
+
+  await memory.saveMessages({
+    messages: [userMessage, assistantMessage],
+    observabilityContext: { tracingContext: { currentSpan: parentSpan } },
+  });
+  ```
+
+- Added per-record config overrides for observation and reflection thresholds in Observational Memory. Each thread can now have its own `messageTokens` and `observationTokens` thresholds that override the instance-level defaults, without requiring a process restart or cache invalidation. If no per-record override is set, the instance-level config is used as before. ([#15102](https://github.com/mastra-ai/mastra/pull/15102))
 
 ### Patch Changes
 
 - Fixed observational memory buffering so sealed assistant chunks stay split instead of being merged back into one persisted message during long tool runs. ([#14995](https://github.com/mastra-ai/mastra/pull/14995))
 
-- Updated dependencies [[`e133cc7`](https://github.com/mastra-ai/mastra/commit/e133cc73c2a73d90a39ad73f724e9cb492466b41), [`c55b527`](https://github.com/mastra-ai/mastra/commit/c55b52758a31368b2077b0dbbc3badfe4063f560), [`e133cc7`](https://github.com/mastra-ai/mastra/commit/e133cc73c2a73d90a39ad73f724e9cb492466b41), [`e133cc7`](https://github.com/mastra-ai/mastra/commit/e133cc73c2a73d90a39ad73f724e9cb492466b41), [`7eb2596`](https://github.com/mastra-ai/mastra/commit/7eb25960d607e07468c9a10c5437abd2deaf1e9a)]:
-  - @mastra/core@0.0.0-studio-deploy-20260404184540
+- Fixed recall() to hide dynamic system reminder messages by default, with includeSystemReminders available when callers need raw reminder history. ([#15100](https://github.com/mastra-ai/mastra/pull/15100))
+
+- Updated dependencies [[`f32b9e1`](https://github.com/mastra-ai/mastra/commit/f32b9e115a3c754d1c8cfa3f4256fba87b09cfb7), [`7d6f521`](https://github.com/mastra-ai/mastra/commit/7d6f52164d0cca099f0b07cb2bba334360f1c8ab), [`a50d220`](https://github.com/mastra-ai/mastra/commit/a50d220b01ecbc5644d489a3d446c3bd4ab30245), [`665477b`](https://github.com/mastra-ai/mastra/commit/665477bc104fd52cfef8e7610d7664781a70c220), [`4cc2755`](https://github.com/mastra-ai/mastra/commit/4cc2755a7194cb08720ff2ab4dffb4b4a5103dfd), [`ac7baf6`](https://github.com/mastra-ai/mastra/commit/ac7baf66ef1db15e03975ef4ebb02724f015a391), [`ed425d7`](https://github.com/mastra-ai/mastra/commit/ed425d78e7c66cbda8209fee910856f98c6c6b82), [`1371703`](https://github.com/mastra-ai/mastra/commit/1371703835080450ef3f9aea58059a95d0da2e5a), [`0df8321`](https://github.com/mastra-ai/mastra/commit/0df832196eeb2450ab77ce887e8553abdd44c5a6), [`98f8a8b`](https://github.com/mastra-ai/mastra/commit/98f8a8bdf5761b9982f3ad3acbe7f1cc3efa71f3), [`ba6f7e9`](https://github.com/mastra-ai/mastra/commit/ba6f7e9086d8281393f2acae60fda61de3bff1f9), [`7eb2596`](https://github.com/mastra-ai/mastra/commit/7eb25960d607e07468c9a10c5437abd2deaf1e9a), [`1805ddc`](https://github.com/mastra-ai/mastra/commit/1805ddc9c9b3b14b63749735a13c05a45af43a80), [`fff91cf`](https://github.com/mastra-ai/mastra/commit/fff91cf914de0e731578aacebffdeebef82f0440), [`61109b3`](https://github.com/mastra-ai/mastra/commit/61109b34feb0e38d54bee4b8ca83eb7345b1d557), [`33f1ead`](https://github.com/mastra-ai/mastra/commit/33f1eadfa19c86953f593478e5fa371093b33779)]:
+  - @mastra/core@1.23.0
+
+## 1.14.0-alpha.2
+
+### Minor Changes
+
+- Added tracing support to Memory operations (recall, save, delete, update working memory). When an `observabilityContext` is provided, Memory creates `MEMORY_OPERATION` spans that capture operation type, message counts, embedding token usage, and vector result counts. Tracing is fully opt-in — existing usage without `observabilityContext` is unaffected. ([#14305](https://github.com/mastra-ai/mastra/pull/14305))
+
+  **Example usage:**
+
+  ```typescript
+  import { Memory } from '@mastra/memory';
+  import { InMemoryStore } from '@mastra/core/storage';
+
+  const memory = new Memory({ storage: new InMemoryStore() });
+
+  // Pass observabilityContext to create observable spans
+  await memory.recall({
+    threadId: 'thread-1',
+    observabilityContext: { tracingContext: { currentSpan: parentSpan } },
+  });
+
+  await memory.saveMessages({
+    messages: [userMessage, assistantMessage],
+    observabilityContext: { tracingContext: { currentSpan: parentSpan } },
+  });
+  ```
+
+- Added per-record config overrides for observation and reflection thresholds in Observational Memory. Each thread can now have its own `messageTokens` and `observationTokens` thresholds that override the instance-level defaults, without requiring a process restart or cache invalidation. If no per-record override is set, the instance-level config is used as before. ([#15102](https://github.com/mastra-ai/mastra/pull/15102))
+
+### Patch Changes
+
+- Fixed recall() to hide dynamic system reminder messages by default, with includeSystemReminders available when callers need raw reminder history. ([#15100](https://github.com/mastra-ai/mastra/pull/15100))
+
+- Updated dependencies [[`ac7baf6`](https://github.com/mastra-ai/mastra/commit/ac7baf66ef1db15e03975ef4ebb02724f015a391), [`0df8321`](https://github.com/mastra-ai/mastra/commit/0df832196eeb2450ab77ce887e8553abdd44c5a6), [`61109b3`](https://github.com/mastra-ai/mastra/commit/61109b34feb0e38d54bee4b8ca83eb7345b1d557), [`33f1ead`](https://github.com/mastra-ai/mastra/commit/33f1eadfa19c86953f593478e5fa371093b33779)]:
+  - @mastra/core@1.23.0-alpha.8
+
+## 1.14.0-alpha.1
+
+### Minor Changes
+
+- Added usage data to ObserveHooks callbacks and standalone reflect() return. ([#15047](https://github.com/mastra-ai/mastra/pull/15047))
+
+  **ObserveHooks:** `onObservationEnd` and `onReflectionEnd` now receive a result object containing token usage from the underlying LLM call. This enables reliable usage tracking across all observation and reflection paths (sync, async buffered, and resource-scoped).
+
+  **Standalone reflect():** `reflect()` now returns `{ reflected, record, usage? }` so callers can capture token usage without hooks.
+
+  **Examples**
+
+  ```ts
+  // Via hooks
+  await memory.observe({
+    threadId,
+    messages,
+    hooks: {
+      onObservationEnd: ({ usage }) => {
+        // usage: { inputTokens, outputTokens, totalTokens }
+      },
+      onReflectionEnd: ({ usage }) => {
+        // usage: { inputTokens, outputTokens, totalTokens }
+      },
+    },
+  });
+
+  // Via standalone reflect()
+  const { reflected, usage } = await memory.reflect(threadId, resourceId);
+  ```
+
+  Existing callbacks that accept no arguments continue to work without changes.
+
+### Patch Changes
+
+- Updated dependencies [[`fff91cf`](https://github.com/mastra-ai/mastra/commit/fff91cf914de0e731578aacebffdeebef82f0440)]:
+  - @mastra/core@1.23.0-alpha.4
 
 ## 1.13.2-alpha.0