chore(deps): bump the prod-dependencies group with 7 updates

[dependabot]

Bumps the prod-dependencies group with 7 updates:

Package	From	To
@cap-js/mcp	1.4.1	1.4.3
@langchain/anthropic	1.5.4	1.5.6
@langchain/core	1.2.5	1.2.8
@langchain/langgraph	1.4.9	1.4.10
deepagents	1.12.2	1.12.4
langchain	1.5.5	1.5.9
@opentelemetry/exporter-trace-otlp-proto	0.57.2	0.221.0

Updates @cap-js/mcp from 1.4.1 to 1.4.3

Release notes

Sourced from @​cap-js/mcp's releases.

v1.4.3

Fixed

• Correctly access target name when executing generic read CQL with infix filters

v1.4.2

Fixed

• Use format: cql as default if the cds.mcp.format option is not applied
• Entities from MCP-enabled services are not shown on index page
• Entity @requires and action @restrict annotations are now respected when filtering tool context

Removed

• Removed the toon_format flag, toon is now always the output format

Changelog

Sourced from @​cap-js/mcp's changelog.

Version 1.4.3 - 2026-08-14

Fixed

• Correctly access target name when executing generic read CQL with infix filters

Version 1.4.2 - 2026-08-13

Fixed

• Use format: cql as default if the cds.mcp.format option is not applied
• Entities from MCP-enabled services are not shown on index page
• Entity @requires and action @restrict annotations are now respected when filtering tool context

Removed

• Removed the toon_format flag, toon is now always the output format

Commits

• af059d6 fix: correctly access target entity when executing query (#60)
• 142c99c chore: prep v1.4.2 (#57)
• 0571b1f fix: entity + service auth filter (#56)
• 9b46b70 fix: remove entities from index with interceptor stack (#55)
• 49d2862 chore: remove toon flag (#54)
• 3bd1764 refactor: split tools.js in smaller files (#52)
• d18695d . (#53)
• 5ec7334 fix: cql as default w/o config (#51)
• b143da2 adjust. (#48)
• 9c478e5 docs: adjust README.md (#49)
• Additional commits viewable in compare view

Updates @langchain/anthropic from 1.5.4 to 1.5.6

Release notes

Sourced from @​langchain/anthropic's releases.

@​langchain/anthropic@​1.5.6

Patch Changes

• #11342 3b0e4c4 Thanks @​thushanth-bengre-langchain! - feat(anthropic): mark Anthropic provider errors as retryable or not

  Builds on stampRetryable in @langchain/core so the retry middleware can tell a transient failure from a deterministic one. Rate limits are marked retryable; context overflow, invalid tool results, bad credentials, and unknown models non-retryable. Anything else stays unmarked and retries as before.

  Also forwards a per-call maxRetries to the retry loop, so a surrounding retry loop such as modelRetryMiddleware can take over instead of the two multiplying against each other.

  Errors keep their original class, so instanceof against the @anthropic-ai/sdk error types is unaffected.

@​langchain/anthropic@​1.5.5

Patch Changes

• #11359 d36ec6a Thanks @​talarari! - fix(anthropic): preserve gateway cost on the native stream path

  convertAnthropicStream now surfaces an Anthropic-compatible gateway's numeric usage.cost at response_metadata.usage.cost, matching the chunk path. Token accounting in usage_metadata is unchanged.

Commits

• d0c905d chore: version packages (#11370)
• 8ba1fd9 fix(aws): classify Bedrock stream-idle timeouts, cover pre-response hang (#11...
• 3b0e4c4 feat(core): mark errors as retryable or not, and stop retrying the ones that ...
• d6ad973 fix(langchain): use unified endpoint for gateway (#11369)
• 3fedf7f chore: version packages (#11367)
• c068bbf fix(core,langchain): patch and release core, update peer dependencies (#11366)
• 242b986 chore: version packages (#11361)
• 2d1f744 chore(langchain): add missing changeset (#11363)
• f401396 feat(langchain): add langsmith gateway to initChatModel (#11362)
• d36ec6a fix(anthropic): preserve gateway cost on the native stream path (#11359)
• Additional commits viewable in compare view

Updates @langchain/core from 1.2.5 to 1.2.8

Release notes

Sourced from @​langchain/core's releases.

@​langchain/core@​1.2.8

Patch Changes

• #11369 d6ad973 Thanks @​hntrl! - fix(langchain): use unified endpoint for gateway

• #11342 3b0e4c4 Thanks @​thushanth-bengre-langchain! - feat(core): mark errors as retryable or not, and stop retrying the ones that aren't

  Retry middleware retried every failure up to maxRetries, including deterministic ones like a bad API key or an unknown model. Retries also nest, so a single such failure could cost dozens of API calls.

  @langchain/core/errors adds stampRetryable(error, retryable) and getRetryable(error). Marking an error leaves its class and shape untouched, so a provider SDK error can be classified without breaking instanceof. getRetryable returns undefined for errors nobody classified, and both are exported so tool authors can mark their own failures.

  modelRetryMiddleware and toolRetryMiddleware now respect the mark by default, and retries stop as soon as one is found rather than each layer spending its own budget. Aborted calls, context overflow, and oversized payloads are marked non-retryable out of the box. Models accept a per-call maxRetries so a surrounding retry loop can take over.

  Behavior change: errors marked non-retryable now fail on the first attempt. Unclassified errors — including any from third-party integrations or custom tools — retry exactly as before. Pass retryOn: () => true to restore the old default. A custom onFailedAttempt replaces the built-in handler and opts out of marking.

@​langchain/core@​1.2.7

Patch Changes

• #11366 c068bbf Thanks @​hntrl! - fix(core,langchain): patch and release core, update peer dependencies

@​langchain/core@​1.2.6

Patch Changes

• #11344 f08e0c6 Thanks @​hntrl! - fix: apply [Symbol.hasInstance] method to all comparable properties using .isInstance()

  We have some internal schemas that rely on z.instanceof(). This uses a strict instanceof check which can conflict if there are multiple versions of core installed. This overrides the Symbol.hasInstance method to use the same logic as .isInstance() to compare objects at runtime.

Commits

• d0c905d chore: version packages (#11370)
• 8ba1fd9 fix(aws): classify Bedrock stream-idle timeouts, cover pre-response hang (#11...
• 3b0e4c4 feat(core): mark errors as retryable or not, and stop retrying the ones that ...
• d6ad973 fix(langchain): use unified endpoint for gateway (#11369)
• 3fedf7f chore: version packages (#11367)
• c068bbf fix(core,langchain): patch and release core, update peer dependencies (#11366)
• 242b986 chore: version packages (#11361)
• 2d1f744 chore(langchain): add missing changeset (#11363)
• f401396 feat(langchain): add langsmith gateway to initChatModel (#11362)
• d36ec6a fix(anthropic): preserve gateway cost on the native stream path (#11359)
• Additional commits viewable in compare view

Updates @langchain/langgraph from 1.4.9 to 1.4.10

Release notes

Sourced from @​langchain/langgraph's releases.

@​langchain/langgraph@​1.4.10

Patch Changes

• #2690 00f68a1 Thanks @​saad-supports-langchain! - fix(langgraph): keep context values out of tracer-derived metadata

Changelog

Sourced from @​langchain/langgraph's changelog.

1.4.10

Patch Changes

• #2690 00f68a1 Thanks @​saad-supports-langchain! - fix(langgraph): keep context values out of tracer-derived metadata

Commits

• f6b41bf chore: version packages (#2698)
• 00f68a1 fix(langgraph): keep context values out of tracer-derived metadata (#2690)
• 6ac60da chore(deps): bump the langchain group across 1 directory with 5 updates (#2651)
• See full diff in compare view

Updates deepagents from 1.12.2 to 1.12.4

Release notes

Sourced from deepagents's releases.

deepagents@1.12.4

Patch Changes

• #747 b2afb8d Thanks @​casparb! - fix(deepagents): batch concurrent Context Hub mutations

• #751 68337fc Thanks @​thushanth-bengre-langchain! - fix(deepagents): coerce grep tool's max_count to a number

• #749 7550c65 Thanks @​thushanth-bengre-langchain! - fix(deepagents): exclude summarization state from subagent input/output

deepagents@1.12.3

Patch Changes

• #724 77e104f Thanks @​gethin-langchain! - feat(deepagents): add output_mode parameter to the grep tool (files_with_matches / content / count)

• #732 239be7e Thanks @​hntrl! - fix(deepagents): disable summary-input trimming by default

  Match Python DeepAgents by providing the full selected conversation to the summarizer unless trimTokensToSummarize is explicitly configured. This prevents oversized tool results from producing context-empty summaries under the default configuration.

• #739 1439bbf Thanks @​taoche! - fix(deepagents): extract text from content blocks when building the summary

Commits

• 2a01d04 chore: version packages (#750)
• 68337fc fix(deepagents): coerce grep tool's max_count to a number (#751)
• 7550c65 fix(deepagents): exclude summarization state from subagent input/output (#749)
• b2afb8d fix(deepagents): batch concurrent Context Hub mutations (#747)
• 44d6d3d chore: version packages (#733)
• 1439bbf fix(deepagents): extract text from content blocks when building the summary (...
• 77e104f feat(deepagents): add output_mode parameter to the grep tool (#724)
• 239be7e fix(deepagents): disable summary input trimming by default (#732)
• 7398fe1 build(deps): bump the patch-deps-updates-main group across 1 directory with 1...
• 792c627 build(deps): bump the minor-deps-updates-main group with 6 updates (#721)
• Additional commits viewable in compare view

Updates langchain from 1.5.5 to 1.5.9

Release notes

Sourced from langchain's releases.

langchain@1.5.9

Patch Changes

• #11369 d6ad973 Thanks @​hntrl! - fix(langchain): use unified endpoint for gateway

• #11342 3b0e4c4 Thanks @​thushanth-bengre-langchain! - feat(core): mark errors as retryable or not, and stop retrying the ones that aren't

  Retry middleware retried every failure up to maxRetries, including deterministic ones like a bad API key or an unknown model. Retries also nest, so a single such failure could cost dozens of API calls.

  @langchain/core/errors adds stampRetryable(error, retryable) and getRetryable(error). Marking an error leaves its class and shape untouched, so a provider SDK error can be classified without breaking instanceof. getRetryable returns undefined for errors nobody classified, and both are exported so tool authors can mark their own failures.

  modelRetryMiddleware and toolRetryMiddleware now respect the mark by default, and retries stop as soon as one is found rather than each layer spending its own budget. Aborted calls, context overflow, and oversized payloads are marked non-retryable out of the box. Models accept a per-call maxRetries so a surrounding retry loop can take over.

  Behavior change: errors marked non-retryable now fail on the first attempt. Unclassified errors — including any from third-party integrations or custom tools — retry exactly as before. Pass retryOn: () => true to restore the old default. A custom onFailedAttempt replaces the built-in handler and opts out of marking.

langchain@1.5.8

Patch Changes

• #11366 c068bbf Thanks @​hntrl! - fix(core,langchain): patch and release core, update peer dependencies

langchain@1.5.7

Patch Changes

• #11363 2d1f744 Thanks @​hntrl! - feat(langchain): add langsmith gateway to initChatModel- #11362

langchain@1.5.6

Patch Changes

• #11331 18765b0 Thanks @​thushanth-bengre-langchain! - fix(langchain): exclude middleware-internal model calls from the message projection

  Bookkeeping model calls made by summarizationMiddleware and toolEmulatorMiddleware no longer appear in run.messages or stream({ streamMode: "messages" }), and the summary summarizationMiddleware writes back to state is no longer projected as a new message. These calls remain observable via streamEvents({ version: "v2" }), identified by lc_source.

• #11344 f08e0c6 Thanks @​hntrl! - fix: apply [Symbol.hasInstance] method to all comparable properties using .isInstance()

  We have some internal schemas that rely on z.instanceof(). This uses a strict instanceof check which can conflict if there are multiple versions of core installed. This overrides the Symbol.hasInstance method to use the same logic as .isInstance() to compare objects at runtime.

Commits

• d0c905d chore: version packages (#11370)
• 8ba1fd9 fix(aws): classify Bedrock stream-idle timeouts, cover pre-response hang (#11...
• 3b0e4c4 feat(core): mark errors as retryable or not, and stop retrying the ones that ...
• d6ad973 fix(langchain): use unified endpoint for gateway (#11369)
• 3fedf7f chore: version packages (#11367)
• c068bbf fix(core,langchain): patch and release core, update peer dependencies (#11366)
• 242b986 chore: version packages (#11361)
• 2d1f744 chore(langchain): add missing changeset (#11363)
• f401396 feat(langchain): add langsmith gateway to initChatModel (#11362)
• d36ec6a fix(anthropic): preserve gateway cost on the native stream path (#11359)
• Additional commits viewable in compare view

Updates @opentelemetry/exporter-trace-otlp-proto from 0.57.2 to 0.221.0

Release notes

Sourced from @​opentelemetry/exporter-trace-otlp-proto's releases.

experimental/v0.221.0

0.221.0

💥 Breaking Changes

• feat(sdk-logs)!: configure the force flush timeout per call #6931 @​LarryHu0217
  • (user-facing): LoggerProviderOptions.forceFlushTimeoutMillis has been removed; pass timeoutMillis to LoggerProvider.forceFlush() instead.
• feat(instrumentation-http)!: emit only stable HTTP semantic conventions. The OTEL_SEMCONV_STABILITY_OPT_IN environment variable no longer changes HTTP attribute or metric emission — old (v1.7.0) and duplicate (http/http/dup) semconv outputs have been removed. #6819 @​maryliag
• feat(instrumentation-fetch)!: emit only stable HTTP semantic conventions. The semconvStabilityOptIn instrumentation config option has been removed; old (v1.7.0) and duplicate semconv outputs are no longer emitted. #6819 @​maryliag
• feat(instrumentation-xml-http-request)!: emit only stable HTTP semantic conventions. The semconvStabilityOptIn instrumentation config option has been removed; old (v1.7.0) and duplicate semconv outputs are no longer emitted. #6819 @​maryliag
• feat(instrumentation-grpc)!: emit only stable network semantic conventions. The OTEL_SEMCONV_STABILITY_OPT_IN environment variable no longer changes attribute emission — net.peer.name and net.peer.port (old) are no longer set; only server.address and server.port (stable). #6819 @​maryliag

🚀 Features

• feat(sdk-logs): allow modifying ReadWriteLogRecord properties (including hrTime, hrTimeObserved, and spanContext) in accordance with the OpenTelemetry Logs specification #6923 @​Babul422
• feat(sdk-node): emit a deprecation warning when the JaegerPropagator is selected via OTEL_PROPAGATORS or declarative config; use tracecontext instead. @​pichlermarc
• feat(instrumentation-http): set error.type to status code in metrics for error requests. #6919 @​raphael-theriault-swi

experimental/v0.220.0

0.220.0

💥 Breaking Changes

• refactor(sdk-logs)!: refactor BatchLogRecordProcessor constructor signature #6817 @​trentm
  • (user-facing): BatchLogRecordProcessor now takes a single options object with all possible properties, instead of two separate arguments. For example, before new BatchLogRecordProcessor(exporter, { maxQueueSize: 1000 }), after new BatchLogRecordProcessor({ exporter, maxQueueSize: 1000 }).
  • interface BufferConfig -> interface BatchLogRecordProcessorOptions, and now includes the exporter property
  • interface BatchLogRecordProcessorBrowserConfig -> interface BatchLogRecordProcessorBrowserOptions
  • (user-facing): SimpleLogRecordProcessor now takes a single options object with all possible properties. For example, before new SimpleLogRecordProcessor(exporter), after new SimpleLogRecordProcessor({ exporter }). #6836
• refactor(configuration)!: change config file parsing to not add default values, nor merge *_list fields #6765 @​trentm
  • Responsibility for handling declarative config "nullBehavior" and "defaultBehavior" belongs in the "create" handling, currently in the "sdk-node" package.
• docs(shim-opencensus): Notice: The @opentracing/shim-opencensus package will be removed in SDK 3.x, planned for approximately September 2026.
  • The OpenCensus and OpenTracing compatibility requirements in the OpenTelemetry specification have been deprecated.
• chore(sdk-node)!: Drop support for deprecated OpenCensusMetricProducer from declarative config

🚀 Features

• feat(configuration): bump config schema to v1.1.0; rename without_scope_info → scope_info_enabled and without_target_info/development → target_info_enabled/development on the Prometheus pull exporter (semantics inverted), rename with_resource_constant_labels → resource_constant_labels. Validate file_format per the configuration versioning spec: accept any minor version of major 1 (e.g. 1.0, 1.1), warn when the minor version is newer than supported, and reject other major versions. #6781 @​MikeGoldsmith
• feat(sdk-node): wire up id_generator from declarative config #6782 @​MikeGoldsmith
• feat(sdk-node): wire up tracer_provider.sampler from declarative config (always_on, always_off, trace_id_ratio_based, parent_based); unrecognized variants warn and fall back to ParentBased(AlwaysOn) #6506 @​MikeGoldsmith
• feat(propagator-env-carrier): empty name normalization #6827 @​pellared
• feat(propagator-env-carrier): make EnvironmentGetter read the current process.env #6853 @​pellared

🐛 Bug Fixes

• fix(sdk-logs): stop Logger.emit() doing work (record construction, metrics, processor onEmit) after the LoggerProvider has shut down #6826 @​anneheartrecord
• fix(sdk-node): pass all config properties (endpoint, headers, timeout, TLS, compression, temporality preference, default histogram aggregation) to OTLP metric exporters in declarative config #6814 @​MikeGoldsmith
• fix(sdk-logs): default BatchLogRecordProcessor scheduleDelayMillis is 1000 #6796 @​trentm
• fix(configuration): percent-decode keys and values in resource.attributes_list per spec #6787 @​MikeGoldsmith
• fix(configuration): default log_level to info in env-based config initialization for consistency with file-based config #6788 @​MikeGoldsmith

... (truncated)

Commits

• 76fa6b5 chore: prepare next release (#6942)
• 3787896 chore(deps): update dependency webpack-cli to v7.2.1 (#6934)
• be5f757 fix(deps): update dependency body-parser to v2.3.0 [security] (#6941)
• f6d8fbe chore(deps): lock file maintenance (#6559)
• 9612732 chore: remove examples/dice from workspaces (#6937)
• 7107906 chore: start using min-release-age in .npmrc, disable minimumReleaseAge for r...
• a7e5d11 chore(deps): update dependency webpack to v5.108.4 (#6933)
• af7a82d chore(deps): update dependency msw to v2.15.0 (#6831)
• b9f57c6 chore(deps): update dependency @​types/webpack-env to v1.18.8 (#6877)
• cbb4abf chore(deps): update ubuntu docker tag to v26 (#6635)
• Additional commits viewable in compare view

Maintainer changes

This version was pushed to npm by GitHub Actions, a new releaser for @​opentelemetry/exporter-trace-otlp-proto since your current version.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
• @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
• @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
• @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
• @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions