chore(deps): bump the prod-dependencies group across 1 directory with 53 updates

[dependabot]

Bumps the prod-dependencies group with 52 updates in the / directory:

Package	From	To
@faker-js/faker	7.0.1	10.4.0
@keycloak/keycloak-admin-client	23.0.7	26.6.1
@mastra/core	1.12.0	1.25.0
@mastra/libsql	1.7.0	1.8.1
@mastra/memory	1.7.0	1.15.1
@mastra/pg	1.8.0	1.9.1
amqp-connection-manager	4.1.15	5.0.0
amqplib	0.10.9	1.0.3
axios	1.15.0	1.15.2
class-transformer	0.3.1	0.5.1
class-validator	0.14.3	0.15.1
connect-redis	6.1.3	9.0.0
cors	2.8.5	2.8.6
crypto-random-string	3.3.1	5.0.0
csv-stringify	6.6.0	6.7.0
dayjs	1.11.19	1.11.20
dompurify	3.4.0	3.4.1
dotenv	10.0.0	17.4.2
envalid	7.3.1	8.1.1
express	4.22.1	5.2.1
express-http-context	1.2.5	2.0.1
express-rate-limit	7.5.1	8.3.2
express-session	1.18.2	1.19.0
express-validator	6.15.0	7.3.2
flat	5.0.2	6.0.1
hasha	5.2.2	7.0.0
helmet	3.23.3	8.1.0
http-status-codes	1.4.0	2.3.0
ical-generator	3.6.1	10.2.0
json-schema-faker	0.5.9	0.6.1
jwks-rsa	2.1.5	4.0.1
jwt-decode	3.1.2	4.0.0
knex	2.5.1	3.2.9
mkdirp	1.0.4	3.0.1
monaco-editor	0.52.0	0.55.1
mongoose	6.13.8	9.5.0
node-schedule	1.3.3	2.1.1
notifications-node-client	7.0.6	8.3.2
passport	0.6.0	0.7.0
pdf2json	4.0.2	4.0.3
puppeteer	22.15.0	24.42.0
qs	6.14.2	6.15.1
react-error-boundary	4.0.12	6.1.1
redis	3.1.2	5.12.1
rehype-raw	6.1.1	7.0.0
remark-gfm	3.0.1	4.0.1
tar-stream	3.1.7	3.1.8
uuid	9.0.1	14.0.0
valid-filename	3.1.0	4.0.0
validator	13.15.26	13.15.35
worker-timers	8.0.28	8.0.31
zod	3.25.76	4.3.6

Updates @faker-js/faker from 7.0.1 to 10.4.0

Release notes

Sourced from @​faker-js/faker's releases.

v10.4.0

What's Changed

• feat(locale): add Norwegian (nb_NO) country definition by @​TomSchrier in faker-js/faker#3714
• refactor(docs): share refreshable code logic by @​ST-DDT in faker-js/faker#3739
• feat(locale): add Japanese cat breed definitions by @​atzzCokeK in faker-js/faker#3716
• feat(food): add plant-based dish variety by @​stuckvgn in faker-js/faker#3745
• fix(locales): correct typos and capitalization in es_MX street names by @​quserforgitp in faker-js/faker#3737
• feat(locale): add Japanese bear definitions by @​atzzCokeK in faker-js/faker#3720
• feat: fi locale phone numbers by @​andeke07 in faker-js/faker#3747
• refactor(hacker): use helpers.fake() instead of helpers.mustache() in phrase() by @​atzzCokeK in faker-js/faker#3736
• chore(deps): update all non-major dependencies by @​renovate[bot] in faker-js/faker#3752
• chore(deps): update dependency @​vitest/eslint-plugin to v1.6.9 by @​renovate[bot] in faker-js/faker#3749
• chore(deps): update eslint by @​renovate[bot] in faker-js/faker#3751
• feat(locale): add Japanese cattle breed definitions by @​atzzCokeK in faker-js/faker#3717
• feat(locale): add Japanese bird definitions by @​atzzCokeK in faker-js/faker#3719
• feat(locale): add Japanese fish definitions by @​atzzCokeK in faker-js/faker#3721
• chore(deps): lock file maintenance by @​renovate[bot] in faker-js/faker#3738
• chore(deps): update devdependencies by @​renovate[bot] in faker-js/faker#3750
• chore(deps): update all non-major dependencies by @​renovate[bot] in faker-js/faker#3754
• refactor(locale): filter and cleanup PersonEntryDefintions data by @​ST-DDT in faker-js/faker#3266
• feat(locale): add Japanese horse breed definitions by @​atzzCokeK in faker-js/faker#3718
• docs: migrate vitepress from v1 to v2.0.0-alpha.17 by @​Shinigami92 in faker-js/faker#3757
• chore(deps): update devdependencies by @​renovate[bot] in faker-js/faker#3755
• chore(deps): lock file maintenance by @​renovate[bot] in faker-js/faker#3756
• chore(deps): update mcr.microsoft.com/devcontainers/typescript-node:24 docker digest to 3ff0e3f by @​renovate[bot] in faker-js/faker#3762
• chore(deps): update eslint by @​renovate[bot] in faker-js/faker#3763
• chore(deps): update devdependencies by @​renovate[bot] in faker-js/faker#3764
• chore(deps): update vitest by @​renovate[bot] in faker-js/faker#3765
• chore(deps): update pnpm/action-setup action to v5 by @​renovate[bot] in faker-js/faker#3766
• chore(deps): update all non-major dependencies by @​renovate[bot] in faker-js/faker#3767
• chore(deps): lock file maintenance by @​renovate[bot] in faker-js/faker#3758
• chore(release): 10.4.0 by @​fakerjs-bot in faker-js/faker#3768

New Contributors

• @​stuckvgn made their first contribution in faker-js/faker#3745
• @​quserforgitp made their first contribution in faker-js/faker#3737
• @​andeke07 made their first contribution in faker-js/faker#3747

Full Changelog: faker-js/faker@v10.3.0...v10.4.0

v10.3.0

What's Changed

• chore(deps): lock file maintenance by @​renovate[bot] in faker-js/faker#3689
• fix(location): state name to 'Trøndelag' for nb_NO by @​Nilhenrik in faker-js/faker#3691
• fix(locale): remove empty string from Hebrew lorem words by @​erezcor in faker-js/faker#3698
• test: add custom matcher "toStartWith" by @​xDivisionByZerox in faker-js/faker#3700
• feat(person): sexType can return 'generic' by @​ST-DDT in faker-js/faker#3259
• refactor(locale): normalize system locale data by @​xDivisionByZerox in faker-js/faker#3702
• feat(locale): add Japanese suffix definitions for person module by @​atzzCokeK in faker-js/faker#3704
• feat(locale): add Japanese job definitions for person module by @​atzzCokeK in faker-js/faker#3705

... (truncated)

Changelog

Sourced from @​faker-js/faker's changelog.

10.4.0 (2026-03-23)

New Locales

• locale: add Japanese bear definitions (#3720) (2a4b15c)
• locale: add Japanese bird definitions (#3719) (dc31ff8)
• locale: add Japanese cat breed definitions (#3716) (54af8a8)
• locale: add Japanese cattle breed definitions (#3717) (c2c7342)
• locale: add Japanese fish definitions (#3721) (15fc361)
• locale: add Japanese horse breed definitions (#3718) (e02536e)
• locale: add Norwegian (nb_NO) country definition (#3714) (614b4e9)

Features

• fi locale phone numbers (#3747) (7afa8b5)
• food: add plant-based dish variety (#3745) (41edf49)

Changed Locales

• locale: filter and cleanup PersonEntryDefintions data (#3266) (67defc8)

Bug Fixes

• locales: correct typos and capitalization in es_MX street names (#3737) (2b32c28)

10.3.0 (2026-02-06)

New Locales

• locale: add Japanese dog definition (#3715) (76c9df1)
• locale: add Japanese color definitions (#3707) (bbbb215)
• locale: add Japanese food module (#3706) (71d55c0)
• locale: add Japanese internet definitions (#3708) (184a709)
• locale: add Japanese job definitions for person module (#3705) (e7f3ccd)
• locale: add Japanese suffix definitions for person module (#3704) (45ad7d8)
• locale: add Norwegian (nb_NO) continent definitions (#3712) (c0f0f23)
• locale: add Norwegian (nb_NO) direction definition (#3713) (43b18fa)
• locale: add Norwegian (nb_NO) sex definitions (#3710) (76063f2)
• locale: add Norwegian (nb_NO) vehicle definition (#3732) (d1c32b0)

Features

• locales: add Norwegian (nb_NO) zodiac sign definitions (#3711) (e306542)
• person: sexType can return 'generic' (#3259) (0e099a1)

... (truncated)

Commits

• b8abfc6 chore(release): 10.4.0 (#3768)
• 7108155 chore(deps): lock file maintenance (#3758)
• 5e6cf2b chore(deps): update all non-major dependencies (#3767)
• 91c944b chore(deps): update pnpm/action-setup action to v5 (#3766)
• cb18595 chore(deps): update vitest (#3765)
• af25d6b chore(deps): update devdependencies (#3764)
• 2e72c27 chore(deps): update eslint (#3763)
• 9a18091 chore(deps): update mcr.microsoft.com/devcontainers/typescript-node:24 docker...
• aa7b6c0 chore(deps): lock file maintenance (#3756)
• 89ba345 chore(deps): update devdependencies (#3755)
• Additional commits viewable in compare view

Maintainer changes

This version was pushed to npm by GitHub Actions, a new releaser for @​faker-js/faker since your current version.

Updates @keycloak/keycloak-admin-client from 23.0.7 to 26.6.1

Release notes

Sourced from @​keycloak/keycloak-admin-client's releases.

26.6.1

26.6.0

... (truncated)

Commits

• 5bebfac Set version to 26.6.1
• d4fbe3f removed postinstall script (#47907) (#47998)
• db330f5 Bump lodash-es from 4.17.23 to 4.18.1 in /js (#47729)
• 00c0dee Display Javascript policy description and code in admin UI
• 3166ff8 Update the OpenAPI spec in the JS module (#47602)
• 3be2527 Reflect Client v2 UUIDs in Operator (#47301)
• d088ee6 go back to using community version of kiota gen (#47264)
• 9ccf47b fix(admin-v2): add description to OpenAPI schema properties (#47302)
• d11136f Separate password and OTP brute force protection to prevent OTP bypass attack...
• de3d854 Add suport for filtering Components by providerId
• Additional commits viewable in compare view

Updates @mastra/core from 1.12.0 to 1.25.0

Release notes

Sourced from @​mastra/core's releases.

April 7, 2026

Highlights

End-to-end RAG Tracing + New RAG/Graph Span Types

RAG ingestion and query operations are now visible in Mastra tracing with new span types (e.g., RAG_INGESTION, RAG_EMBEDDING, RAG_VECTOR_OPERATION, RAG_ACTION, GRAPH_ACTION) plus helpers like startRagIngestion() / withRagIngestion(). Instrumentation is opt-in via an observabilityContext, and @mastra/rag automatically threads context from agent TOOL_CALL spans into vector and graph tools.

CloudExporter Now Ships All Observability Signals (Not Just Traces)

@mastra/observability CloudExporter can now batch and upload logs, metrics, scores, and feedback in addition to tracing spans, enabling a single exporter path to Mastra Cloud for all signals. This also changes the endpoint configuration to use a base collector URL and derive publish paths automatically.

Span Filtering to Reduce Observability Noise and Cost

excludeSpanTypes and spanFilter were added to ObservabilityInstanceConfig in both @mastra/core and @mastra/observability, allowing you to drop entire span categories (e.g., MODEL_CHUNK) or apply predicate-based filtering before export—useful for pay-per-span backends.

AI SDK v6 Message Support + Message Interop Helpers

@mastra/core MessageList can now accept AI SDK v6 UI/model messages and project stored messages via messageList.get.all.aiV6.ui(), supporting v6 approval request/response flows. @mastra/ai-sdk adds toAISdkMessages() to load stored Mastra messages into AI SDK v5 or v6 chat UIs.

Better Reliability and Debuggability Across Logs + MCP + Memory

Observability log correlation is fixed so logs inside agent runs carry the active span correlation fields (restoring trace↔log linking), and deepClean() now applies to all signals and better preserves Map/Set/Error detail. MCP tool discovery now retries after reconnectable errors and the MCP server returns spec-correct 404s for stale sessions; memory recall gains more precise browsing (partType, toolName, threadId: "current", anchor paging), and message parts now include createdAt timestamps for accurate part-level timing.

Breaking Changes

• CloudExporter endpoint format changed: configure a base endpoint URL (publisher paths are derived automatically) when using CloudExporter for Mastra Cloud uploads.

Changelog

@​mastra/core@​1.24.0

Minor Changes

• Added excludeSpanTypes and spanFilter options to ObservabilityInstanceConfig for selectively filtering spans before export. Use excludeSpanTypes to drop entire categories of spans by type (e.g., MODEL_CHUNK, MODEL_STEP) or spanFilter for fine-grained predicate-based filtering by attributes, metadata, entity, or any combination. Both options help reduce noise and costs in observability platforms that charge per-span. (#15131)

  excludeSpanTypes example:

  excludeSpanTypes: [SpanType.MODEL_CHUNK, SpanType.MODEL_STEP, SpanType.WORKFLOW_SLEEP];

  spanFilter example:

  spanFilter: span => {
    if (span.type === SpanType.MODEL_CHUNK) return false;
    if (span.type === SpanType.TOOL_CALL && span.attributes?.success) return false;
    return true;
  };

  Resolves mastra-ai/mastra#12710

• Add RAG observability (#10898) (#15137)

  Surfaces RAG ingestion and query operations in Mastra's AI tracing.

... (truncated)

Commits

• ed68de0 chore: version - exit prerelease mode
• d34368b chore: regenerate providers and docs [skip ci]
• d4b2067 chore: version packages
• d7163fe chore: regenerate providers and docs [skip ci]
• 184907d feat(core): upgrade model router providers to v3 spec, fix OpenAI reasoning e...
• 8dcc77e fix(agent-browser): initialize browser for default thread ID in thread scope ...
• 8fe46d3 fix(core): list_files empty pattern + harness resume approval (#15360)
• 18c67db feat(core): unify versions, evals, and observability (#15317)
• fa8140b fix(editor): fork agent instead of mutating singleton in applyStoredOverrides...
• 190f452 feat(browser): add profile and executablePath options (#15194)
• Additional commits viewable in compare view

Updates @mastra/libsql from 1.7.0 to 1.8.1

Release notes

Sourced from @​mastra/libsql's releases.

February 26, 2026

Highlights

Supervisor Pattern for Multi-Agent Coordination

A new supervisor pattern enables orchestrating multiple agents via stream() and generate(), with delegation hooks, iteration monitoring, completion scoring, memory isolation, tool approval propagation, context filtering, and a bail mechanism.

Metadata-Only Vector Queries (Optional queryVector)

Vector querying now supports metadata-only retrieval by making queryVector optional (with at least one of queryVector or filter required). @mastra/pg’s PgVector.query() explicitly supports filter-only queries, while other vector stores now throw a clear MastraError when metadata-only queries aren’t supported.

More Flexible Evaluations with runEvals Target Options

runEvals adds targetOptions to forward execution/run options into agent.generate() or workflow.run.start(), plus per-item startOptions for workflow-specific overrides (e.g., initialState) on each eval datum.

LSP Diagnostics After Workspace Edits

Workspace edit tools (write_file, edit_file, ast_edit) can now surface Language Server Protocol diagnostics immediately after edits (TypeScript, Python/Pyright, Go/gopls, Rust/rust-analyzer, ESLint), helping catch type/lint errors before the next tool call.

New Blaxel Cloud Sandbox Provider

@mastra/blaxel adds a Blaxel cloud sandbox provider, expanding deployment/runtime options for executing workspace tooling in a managed environment.

Breaking Changes

• None noted in this changelog.

Changelog

@​mastra/core@​1.8.0

Minor Changes

• Make queryVector optional in the QueryVectorParams interface to support metadata-only queries. At least one of queryVector or filter must be provided. Not all vector store backends support metadata-only queries — check your store's documentation for details. (#13286)

  Also fixes documentation where the query() parameter was incorrectly named vector instead of queryVector.

• Added targetOptions parameter to runEvals that is forwarded directly to agent.generate() (modern path) or workflow.run.start(). Also added per-item startOptions field to RunEvalsDataItem for per-item workflow options like initialState. (#13366)

  New feature: targetOptions

  Pass agent execution options (e.g. maxSteps, modelSettings, instructions) through to agent.generate(), or workflow run options (e.g. perStep, outputOptions) through to workflow.run.start():

  // Agent - pass modelSettings or maxSteps
  await runEvals({
    data,
    scorers,
    target: myAgent,
    targetOptions: { maxSteps: 5, modelSettings: { temperature: 0 } },
  });
  // Workflow - pass run options
  await runEvals({
  data,
  scorers,
  target: myWorkflow,

... (truncated)

Changelog

Sourced from @​mastra/libsql's changelog.

1.8.1

Patch Changes

• Fixed "column does not exist" errors when using experiment review features on databases created before the review pipeline was introduced. Startup now automatically migrates older experiment tables to the latest schema. (#15304)

• Added entityVersionId, parentEntityVersionId, and rootEntityVersionId columns to observability storage tables (spans, metrics, scores, feedback, logs) for filtering and grouping traces by entity version. Added ALTER TABLE migrations for existing databases. Added targetType, targetId, agentVersion, and status filters to listExperiments, and traceId and status filters to listExperimentResults. (#15317)

• Updated dependencies [87df955, 8fad147, 582644c, cbdf3e1, 8fe46d3, 18c67db, 4ba3bb1, 5d84914, 8dcc77e, aa67fc5, fd2f314, fa8140b, 190f452, e80fead, 0287b64, 7e7bf60, 184907d, 075e91a, 0c4cd13, b16a753]:

  • @​mastra/core@​1.25.0

1.8.1-alpha.1

Patch Changes

• Added entityVersionId, parentEntityVersionId, and rootEntityVersionId columns to observability storage tables (spans, metrics, scores, feedback, logs) for filtering and grouping traces by entity version. Added ALTER TABLE migrations for existing databases. Added targetType, targetId, agentVersion, and status filters to listExperiments, and traceId and status filters to listExperimentResults. (#15317)

• Updated dependencies [cbdf3e1, 8fe46d3, 18c67db, 8dcc77e, aa67fc5, fa8140b, 190f452, 7e7bf60, 184907d, 0c4cd13, b16a753]:

  • @​mastra/core@​1.25.0-alpha.3

1.8.1-alpha.0

Patch Changes

• Fixed "column does not exist" errors when using experiment review features on databases created before the review pipeline was introduced. Startup now automatically migrates older experiment tables to the latest schema. (#15304)

1.8.0

Minor Changes

• Implemented updateObservationalMemoryConfig() in Postgres, LibSQL, and MongoDB storage adapters. This enables per-record config overrides for observational memory thresholds, supporting the new memory.updateObservationalMemoryConfig() API in @mastra/memory. (#15115)

Patch Changes

• Updated dependencies [f32b9e1, 7d6f521, a50d220, 665477b, 4cc2755, ac7baf6, ed425d7, 1371703, 0df8321, 98f8a8b, ba6f7e9, 7eb2596, 1805ddc, fff91cf, 61109b3, 33f1ead]:
  • @​mastra/core@​1.23.0

1.8.0-alpha.0

Minor Changes

• Implemented updateObservationalMemoryConfig() in Postgres, LibSQL, and MongoDB storage adapters. This enables per-record config overrides for observational memory thresholds, supporting the new memory.updateObservationalMemoryConfig() API in @mastra/memory. (#15115)

Patch Changes

• Updated dependencies [a50d220]:
  • @​mastra/core@​1.23.0-alpha.9

1.7.4

... (truncated)

Commits

• ed68de0 chore: version - exit prerelease mode
• d4b2067 chore: version packages
• 18c67db feat(core): unify versions, evals, and observability (#15317)
• 0c12355 chore: version packages
• 6544c97 fix(pg,libsql): add schema backfill for experiment table columns (#15304)
• 826fdb6 chore: version - exit prerelease mode
• 6c3e6c1 chore: version packages (alpha) (#15114)
• aced936 feat(memory): per-record config overrides for observation & reflection thresh...
• 7f79615 feat: add from, to, and offset params to getObservationalMemoryHistory (#15060)
• dcb6eeb chore: version - exit prerelease mode
• Additional commits viewable in compare view

Updates @mastra/memory from 1.7.0 to 1.15.1

Release notes

Sourced from @​mastra/memory's releases.

March 17, 2026

Highlights

AI Gateway Tool Support in the Agentic Loop

@mastra/core now supports AI Gateway tools (e.g. gateway.tools.perplexitySearch()) as provider-executed tools: it infers providerExecuted, merges streamed provider results back into the originating tool calls, and skips local execution when the provider already returned a result.

More Reliable Observational Memory (Cache Stability + “As Of” Retrieval)

Observational memory persistence is more stable via dated message boundary delimiters and chunking, and @mastra/memory adds getObservationsAsOf() to retrieve the exact observation set active at a given message timestamp (useful for replay/debugging and consistent prompting).

MCP Client Diagnostics & Per-Server Control

@mastra/mcp adds per-server operational tooling—reconnectServer(serverName), listToolsetsWithErrors(), and getServerStderr(serverName)—to improve reliability and debugging of MCP stdio/server integrations.

Breaking Changes

• None called out in this changelog.

Changelog

@​mastra/core@​1.14.0

Patch Changes

• Update provider registry and model documentation with latest models and providers (51970b3)

• Added dated message boundary delimiters when activating buffered observations for improved cache stability. (#14367)

• Fixed provider-executed tool calls being saved out of order or without results in memory replay. (Fixes #13762) (#13860)

• Fix generateEmptyFromSchema to accept both string and pre-parsed object JSON schema inputs, recursively initialize nested object properties, and respect default values. Updated WorkingMemoryTemplate type to a discriminated union supporting Record<string, unknown> content for JSON format templates. Removed duplicate private schema generator in the working-memory processor in favor of the shared utility. (#14310)

• Fixed provider-executed tool calls (e.g. Anthropic web_search) being dropped or incorrectly persisted when deferred by the provider. Tool call parts are now persisted in stream order, and deferred tool results are correctly merged back into the originating message. (#14282)

• Fixed replaceString utility to properly escape $ characters in replacement strings. Previously, patterns like $& in the replacement text would be interpreted as regex backreferences instead of literal text. (#14434)

• Fixed tool invocation updates to preserve providerExecuted and providerMetadata from the original tool call when updating to result state. (#14431)

• @mastra/core: patch (#14327)

  Added spanId alongside traceId across user-facing execution results that return tracing identifiers (including agent stream/generate and workflow run results) so integrations can query observability vendors by run root span ID

• Add AI Gateway tool support in the agentic loop. (#14016)

  Gateway tools (e.g., gateway.tools.perplexitySearch()) are provider-executed but, unlike native provider tools (e.g., openai.tools.webSearch()), the LLM provider does not store their results server-side. The agentic loop now correctly infers providerExecuted for these tools, merges streamed provider results with their corresponding tool calls, and skips local execution when a provider result is already present.

  Fixes #13190

• Fixed schema-based working memory typing so workingMemory.schema accepts supported schemas such as Zod and JSON Schema. (#14363)

• Fixed workspace search being wiped when skills refresh. Previously, calling skills.refresh() or triggering a skills re-discovery via maybeRefresh() would clear the entire BM25 search index, including auto-indexed workspace content. Now only skill entries are removed from the index during refresh, preserving workspace search results. (#14287)

• Added client/server body schemas for feedback and scores that omit the timestamp field, allowing it to be set server-side (#14270)

... (truncated)

Changelog

Sourced from @​mastra/memory's changelog.

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)

• Updated dependencies [87df955, 8fad147, 582644c, cbdf3e1, 8fe46d3, 18c67db, 4ba3bb1, 5d84914, 8dcc77e, aa67fc5, fd2f314, fa8140b, 190f452, e80fead, 0287b64, 7e7bf60, 184907d, 075e91a, 2a69802, 0c4cd13, b16a753]:

  • @​mastra/core@​1.25.0
  • @​mastra/schema-compat@​1.2.8

1.15.1-alpha.1

Patch Changes

• Updated dependencies [4ba3bb1, 2a69802]:
  • @​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)

• Updated dependencies [87df955, 075e91a]:

  • @​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)

  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)

• Improved observational memory formatting to use part timestamps when rendering dates and times. (#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)

• Updated dependencies [8db7663, 153e864, 715710d, 378c6c4, 9f91fd5, ba6fa9c]:

  • @​mastra/core@​1.24.0

1.15.0-alpha.3

... (truncated)

Commits

• ed68de0 chore: version - exit prerelease mode
• cbdf3e1 feat: add processAPIError processor method for handling LLM API rejections (#...
• 2e553a0 chore: version packages
• 2a69802 fix(schema-compat): improve provider structured output and tool-call compatib...
• fd2f314 fix: surface upstream structured output processor failures (#15226)
• f8c496f chore: version packages
• 075e91a fix(memory): use duck typing for gateway model detection in OM processor (#15...
• 3543668 chore(deps): update react to ^19.2.5 (