chore(deps): bump the python-minor-patch group with 8 updates
Bumps the python-minor-patch group with 8 updates:
| Package | From | To |
|---|---|---|
| langchain-core | 1.7.6 | 1.8.0 |
| pydantic-ai-slim | 2.66.0 | 2.68.0 |
| langchain | 1.4.9 | 1.5.0 |
| openai | 3.46.0 | 3.50.0 |
| pydantic-ai | 2.66.0 | 2.68.0 |
| pytest-asyncio | 1.4.1.post1 | 1.5.0 |
| ruff | 0.17.4 | 0.17.5 |
| ty | 0.0.99 | 0.0.102 |
Updates langchain-core from 1.7.6 to 1.8.0
Release notes
Sourced from langchain-core's releases.
langchain-core==1.8.0
langchain-core1.8.0 drops support for Python 3.10, which has reached end-of-life. The minimum supported version is now Python 3.11. Users on Python 3.10 will continue to resolvelangchain-core1.7.x.Changes since langchain-core==1.7.7
release(core): 1.8.0 (#41796) feat(core): Remove EOL Python 3.10 support (#41267) feat(core): add decision models (#41765)
langchain-core==1.7.7
Changes since langchain-core==1.7.6
release(core): 1.7.7 (#41776) fix(core): cap nesting depth in
load()(#41742) chore(deps): patch newer-python security lock resolutions (#41585)
Commits
a4b4a7brelease(core): 1.8.0 (#41796)1dd41b2feat(core): Remove EOL Python 3.10 support (#41267)8b8969frefactor(typesafe): deprecate experimental middleware (#41768)3ce507cfeat(typesafe): implement TypeSafeClassifier as a decision model (#41767)b067614feat(langchain): add model router and auto mode middleware (#41766)6bbc934feat(core): add decision models (#41765)c6ec96adocs: update OpenWiki (#41782)1085dbdchore(model-profiles): refresh model profile data (#41781)9b86582docs(langchain): explain Baseten minimum version (#41777)c0344e0release(langchain): 1.4.10 (#41778)- Additional commits viewable in compare view
Updates pydantic-ai-slim from 2.66.0 to 2.68.0
Release notes
Sourced from pydantic-ai-slim's releases.
v2.68.0 (2026-11-11)
What's Changed
⚠️ Compatibility Notes
- Note text that looks like an application frame in user and tool content instead of escaping it by
@DouweMin pydantic/pydantic-ai#11813- Add
OutputTokenLimitExceeded, and fall back to the next model inModelFallbackwhen a response is cut off atmax_tokensby@DouweMin pydantic/pydantic-ai#12407- Ask for approvals and answers inside the run with
RequestInput, putAskUseron the deferred pipeline, and mark deferred calls withkindandanswered_byby@DouweMin pydantic/pydantic-ai#12757- Add
DelegationTasks.discover()and stop one damaged saved task from blockingDelegationTasks.opened()by@mpfaffenbergerin pydantic/pydantic-ai#10140- Require workspace backends to leave detached processes running, and check it in
WorkspaceBackendSuiteby@mpfaffenbergerin pydantic/pydantic-ai#13070- Run
PromptInjectionDefendersemantic classification as durable operations so it works on Temporal by@mpfaffenbergerin pydantic/pydantic-ai#12424- Give sub-agent task-control events the
Eventsuffix and past-participle names, withagent_nameandparent_task_idfields, keeping the old names as deprecated aliases by@mpfaffenbergerin pydantic/pydantic-ai#13074- Spell out abbreviated identifiers in
pydantic_ai_harness, renamingnum_resultstomax_resultsandmax_duration_secstomax_duration_secondsby@mpfaffenbergerin pydantic/pydantic-ai#13075- Rename harness
CapabilityEventfields to*_count,*_name,duration_msandprevious_item, keeping the old names as deprecated aliases by@mpfaffenbergerin pydantic/pydantic-ai#13068- Rename harness
CapabilityEventoutliers to<Subject><PastParticiple>Eventand addSTEP_PERSISTENCE_EVENTSby@mpfaffenbergerin pydantic/pydantic-ai#13065- Add the
RefineOutputharness capability by@mpfaffenbergerin pydantic/pydantic-ai#13120🚀 Features
- Load and save UI adapter conversations through a
ConversationStorewithconversation_storeby@DouweMin pydantic/pydantic-ai#12326- Use the AI Gateway's streamed cost for
GoogleModelandBedrockConverseModelstreams by@ddanielcruzzin pydantic/pydantic-ai#13168- Support running
SubAgentsdelegations to durable delegates as Temporal and DBOS child workflows by@mpfaffenbergerin pydantic/pydantic-ai#12516- Add the
RefineOutputharness capability by@mpfaffenbergerin pydantic/pydantic-ai#13120- Add a per-command
cwdtoWorkspace.runby@mpfaffenbergerin pydantic/pydantic-ai#12642- Require workspace backends to leave detached processes running, and check it in
WorkspaceBackendSuiteby@mpfaffenbergerin pydantic/pydantic-ai#13070- Add
DelegationTasks.discover()and stop one damaged saved task from blockingDelegationTasks.opened()by@mpfaffenbergerin pydantic/pydantic-ai#10140- Bound the wait for the detached process to start in the
WorkspaceBackendSuitereattach rule by@mpfaffenbergerin pydantic/pydantic-ai#13081- Ask for approvals and answers inside the run with
RequestInput, putAskUseron the deferred pipeline, and mark deferred calls withkindandanswered_byby@DouweMin pydantic/pydantic-ai#12757- Add
OutputTokenLimitExceeded, and fall back to the next model inModelFallbackwhen a response is cut off atmax_tokensby@DouweMin pydantic/pydantic-ai#12407- Allow a
CancellationTokenfired from a workflow signal to cancel a Temporal durable run by@DouweMin pydantic/pydantic-ai#7210- Note text that looks like an application frame in user and tool content instead of escaping it by
@DouweMin pydantic/pydantic-ai#11813- Add a
GoogleDecisionsModelbackend for the Gemini Decisions API by@dsfacciniin pydantic/pydantic-ai#13181🐛 Bug Fixes
- Resolve
vendor/modelids through one case-insensitive family map onOpenAICompatibleProvider,LiteLLMProviderandTogetherProviderby@DouweMin pydantic/pydantic-ai#12743- Map unified
thinkingtoreasoning_effortfor Z.AI GLM and DeepSeek V3.2 on Bedrock by@EdAbatiin pydantic/pydantic-ai#11491- Honor
Permissionsremembered approvals inside durable execution by@mpfaffenbergerin pydantic/pydantic-ai#12410- Make
AnthropicModel.count_tokenssend the request's betas and raise its all-deferred toolsUserErrorby@dsfacciniin pydantic/pydantic-ai#13098- Drop the account-age claim from
AnthropicStaleThinkingBlockWarning, since Claude Fable 5.2 and Mythos 5.2 bind thinking blocks on every account by@dsfacciniin pydantic/pydantic-ai#13155- Run
PromptInjectionDefendersemantic classification as durable operations so it works on Temporal by@mpfaffenbergerin pydantic/pydantic-ai#12424- Carry
ModelResponse.canonical_systemthrough the Vercel AI and AG-UI event streams by@DouweMin pydantic/pydantic-ai#13134- Lower a
max_tokensabove a Claude model's maximum output to it instead of sending it by@DouweMin pydantic/pydantic-ai#12427- Report when
TieredCompactioncan't reachtarget_tokensbecause its tiers keep more than the target by@DouweMin pydantic/pydantic-ai#12650- Summarize the tail messages
SummarizingCompaction(keep_user_messages=True)copies displace instead of dropping them by@DouweMin pydantic/pydantic-ai#12652- Complete a Gemini Live turn that ends silently after a tool result by
@DouweMin pydantic/pydantic-ai#13126📦 Dependencies
- Raise the
genai-pricesfloor to 0.2.3 so Anthropic fast mode responses are priced at fast mode rates by@dsfacciniin pydantic/pydantic-ai#13104New Contributors
@EdAbatimade their first contribution in pydantic/pydantic-ai#11491Full Changelog: https://github.com/pydantic/pydantic-ai/compare/v2.67.0...v2.68.0
... (truncated)
Commits
d8bcca9Add aGoogleDecisionsModelbackend for the Gemini Decisions API (#13181)8d7ed03Accept prose before the closing JSON fence inPR Review Checksanswers (#13...c429ff0[voice-authority] Note text that looks like an application frame in user and ...aefc43aSummarize the tail messagesSummarizingCompaction(keep_user_messages=True)...96fa403Report whenTieredCompactioncan't reachtarget_tokensbecause its tiers ...3dc712aAllow aCancellationTokenfired from a workflow signal to cancel a Temporal...86727dfComplete a Gemini Live turn that ends silently after a tool result (#13126)03fd4e3EnableWebFetchToolin the gh-aw shim only when theANTHROPIC_BASE_URLho...7ae4e2cLower amax_tokensabove a Claude model's maximum output to it instead of s...fda76d3AddOutputTokenLimitExceeded, and fall back to the next model in `ModelFall...- Additional commits viewable in compare view
Updates langchain from 1.4.9 to 1.5.0
Release notes
Sourced from langchain's releases.
langchain==1.5.0
langchain1.5.0 drops support for Python 3.10, which has reached end-of-life. The minimum supported version is now Python 3.11, and the minimumlangchain-coreversion is now 1.8.0. Users on Python 3.10 will continue to resolvelangchain1.4.x.Changes since langchain==1.4.10
release(langchain): 1.5.0 (#41798) fix(langchain): mark
ModelRouterMiddlewareas beta (#41797) feat(langchain): Remove EOL'd Python 3.10 from langchain_v1 (#41301) feat(langchain): add model router and auto mode middleware (#41766) docs(langchain): explain Baseten minimum version (#41777)langchain==1.4.10
Changes since langchain==1.4.9
release(langchain): 1.4.10 (#41778) fix(langchain): keep middleware-internal model calls off the
messagesstream (#41579) chore(deps): patch newer-python security lock resolutions (#41585) fix(langchain): omit trace inputs fromModelRefusalMiddleware(#41689)
Commits
c2f4036release(langchain): 1.5.0 (#41798)42507d3fix(langchain): markModelRouterMiddlewareas beta (#41797)c60c281feat(langchain): Remove EOL'd Python 3.10 from langchain_v1 (#41301)a4b4a7brelease(core): 1.8.0 (#41796)1dd41b2feat(core): Remove EOL Python 3.10 support (#41267)8b8969frefactor(typesafe): deprecate experimental middleware (#41768)3ce507cfeat(typesafe): implement TypeSafeClassifier as a decision model (#41767)b067614feat(langchain): add model router and auto mode middleware (#41766)6bbc934feat(core): add decision models (#41765)c6ec96adocs: update OpenWiki (#41782)- Additional commits viewable in compare view
Updates openai from 3.46.0 to 3.50.0
Release notes
Sourced from openai's releases.
v3.50.0
3.50.0 (2026-11-12)
Features
Chores
Documentation
v3.49.0
3.49.0 (2026-11-10)
Features
Bug Fixes
v3.48.0
3.48.0 (2026-11-09)
Features
- api: add stream options for agent sessions (#4251) (c6fc264)
- api: recognize gpt-6-rosalind in beta Responses (#4257) (3774b7e)
Chores
- api: retain agent scheduling support (#4255) (a1175a3)
- deps: bump locked multidict from 6.9.1 to 7.1.0 (#4259) (28cdb1b)
- deps-dev: bump the python-maintenance group with 5 updates (#4258) (4b10e22)
- deps-dev: bump types-requests from 2.33.0.20261006 to 2.33.0.20261030 (#4260) (de498d7)
Documentation
... (truncated)
Changelog
Sourced from openai's changelog.
3.50.0 (2026-11-12)
Features
Chores
Documentation
3.49.0 (2026-11-10)
Features
Bug Fixes
3.48.0 (2026-11-09)
Features
- api: add stream options for agent sessions (#4251) (c6fc264)
- api: recognize gpt-6-rosalind in beta Responses (#4257) (3774b7e)
Chores
- api: retain agent scheduling support (#4255) (a1175a3)
- deps: bump locked multidict from 6.9.1 to 7.1.0 (#4259) (28cdb1b)
- deps-dev: bump the python-maintenance group with 5 updates (#4258) (4b10e22)
- deps-dev: bump types-requests from 2.33.0.20261006 to 2.33.0.20261030 (#4260) (de498d7)
Documentation
... (truncated)
Commits
f44c1ecrelease: 3.50.0 (#4265)734f75bfeat(api): expose reasoning input transformations on responses (#4268)141a89adocs: simplify review handoff instructions (#4266)0ebd006chore(deps): bump astral-sh/setup-uv from 10.3.0 to 10.3.3 (#4261)3d573c0release: 3.49.0 (#4263)ea07983feat(api): add typed async function tools (#4264)13c000dfix: encode microphone recordings as valid PCM WAV (#3900)7e1582crelease: 3.48.0 (#4252)de498d7chore(deps-dev): bump types-requests from 2.33.0.20261006 to 2.33.0.20261030 ...28cdb1bchore(deps): bump multidict from 6.9.1 to 7.1.0 (#4259)- Additional commits viewable in compare view
Updates pydantic-ai from 2.66.0 to 2.68.0
Release notes
Sourced from pydantic-ai's releases.
v2.68.0 (2026-11-11)
What's Changed
⚠️ Compatibility Notes
- Note text that looks like an application frame in user and tool content instead of escaping it by
@DouweMin pydantic/pydantic-ai#11813- Add
OutputTokenLimitExceeded, and fall back to the next model inModelFallbackwhen a response is cut off atmax_tokensby@DouweMin pydantic/pydantic-ai#12407- Ask for approvals and answers inside the run with
RequestInput, putAskUseron the deferred pipeline, and mark deferred calls withkindandanswered_byby@DouweMin pydantic/pydantic-ai#12757- Add
DelegationTasks.discover()and stop one damaged saved task from blockingDelegationTasks.opened()by@mpfaffenbergerin pydantic/pydantic-ai#10140- Require workspace backends to leave detached processes running, and check it in
WorkspaceBackendSuiteby@mpfaffenbergerin pydantic/pydantic-ai#13070- Run
PromptInjectionDefendersemantic classification as durable operations so it works on Temporal by@mpfaffenbergerin pydantic/pydantic-ai#12424- Give sub-agent task-control events the
Eventsuffix and past-participle names, withagent_nameandparent_task_idfields, keeping the old names as deprecated aliases by@mpfaffenbergerin pydantic/pydantic-ai#13074- Spell out abbreviated identifiers in
pydantic_ai_harness, renamingnum_resultstomax_resultsandmax_duration_secstomax_duration_secondsby@mpfaffenbergerin pydantic/pydantic-ai#13075- Rename harness
CapabilityEventfields to*_count,*_name,duration_msandprevious_item, keeping the old names as deprecated aliases by@mpfaffenbergerin pydantic/pydantic-ai#13068- Rename harness
CapabilityEventoutliers to<Subject><PastParticiple>Eventand addSTEP_PERSISTENCE_EVENTSby@mpfaffenbergerin pydantic/pydantic-ai#13065- Add the
RefineOutputharness capability by@mpfaffenbergerin pydantic/pydantic-ai#13120🚀 Features
- Load and save UI adapter conversations through a
ConversationStorewithconversation_storeby@DouweMin pydantic/pydantic-ai#12326- Use the AI Gateway's streamed cost for
GoogleModelandBedrockConverseModelstreams by@ddanielcruzzin pydantic/pydantic-ai#13168- Support running
SubAgentsdelegations to durable delegates as Temporal and DBOS child workflows by@mpfaffenbergerin pydantic/pydantic-ai#12516- Add the
RefineOutputharness capability by@mpfaffenbergerin pydantic/pydantic-ai#13120- Add a per-command
cwdtoWorkspace.runby@mpfaffenbergerin pydantic/pydantic-ai#12642- Require workspace backends to leave detached processes running, and check it in
WorkspaceBackendSuiteby@mpfaffenbergerin pydantic/pydantic-ai#13070- Add
DelegationTasks.discover()and stop one damaged saved task from blockingDelegationTasks.opened()by@mpfaffenbergerin pydantic/pydantic-ai#10140- Bound the wait for the detached process to start in the
WorkspaceBackendSuitereattach rule by@mpfaffenbergerin pydantic/pydantic-ai#13081- Ask for approvals and answers inside the run with
RequestInput, putAskUseron the deferred pipeline, and mark deferred calls withkindandanswered_byby@DouweMin pydantic/pydantic-ai#12757- Add
OutputTokenLimitExceeded, and fall back to the next model inModelFallbackwhen a response is cut off atmax_tokensby@DouweMin pydantic/pydantic-ai#12407- Allow a
CancellationTokenfired from a workflow signal to cancel a Temporal durable run by@DouweMin pydantic/pydantic-ai#7210- Note text that looks like an application frame in user and tool content instead of escaping it by
@DouweMin pydantic/pydantic-ai#11813- Add a
GoogleDecisionsModelbackend for the Gemini Decisions API by@dsfacciniin pydantic/pydantic-ai#13181🐛 Bug Fixes
- Resolve
vendor/modelids through one case-insensitive family map onOpenAICompatibleProvider,LiteLLMProviderandTogetherProviderby@DouweMin pydantic/pydantic-ai#12743- Map unified
thinkingtoreasoning_effortfor Z.AI GLM and DeepSeek V3.2 on Bedrock by@EdAbatiin pydantic/pydantic-ai#11491- Honor
Permissionsremembered approvals inside durable execution by@mpfaffenbergerin pydantic/pydantic-ai#12410- Make
AnthropicModel.count_tokenssend the request's betas and raise its all-deferred toolsUserErrorby@dsfacciniin pydantic/pydantic-ai#13098- Drop the account-age claim from
AnthropicStaleThinkingBlockWarning, since Claude Fable 5.2 and Mythos 5.2 bind thinking blocks on every account by@dsfacciniin pydantic/pydantic-ai#13155- Run
PromptInjectionDefendersemantic classification as durable operations so it works on Temporal by@mpfaffenbergerin pydantic/pydantic-ai#12424- Carry
ModelResponse.canonical_systemthrough the Vercel AI and AG-UI event streams by@DouweMin pydantic/pydantic-ai#13134- Lower a
max_tokensabove a Claude model's maximum output to it instead of sending it by@DouweMin pydantic/pydantic-ai#12427- Report when
TieredCompactioncan't reachtarget_tokensbecause its tiers keep more than the target by@DouweMin pydantic/pydantic-ai#12650- Summarize the tail messages
SummarizingCompaction(keep_user_messages=True)copies displace instead of dropping them by@DouweMin pydantic/pydantic-ai#12652- Complete a Gemini Live turn that ends silently after a tool result by
@DouweMin pydantic/pydantic-ai#13126📦 Dependencies
- Raise the
genai-pricesfloor to 0.2.3 so Anthropic fast mode responses are priced at fast mode rates by@dsfacciniin pydantic/pydantic-ai#13104New Contributors
@EdAbatimade their first contribution in pydantic/pydantic-ai#11491Full Changelog: https://github.com/pydantic/pydantic-ai/compare/v2.67.0...v2.68.0
... (truncated)
Commits
d8bcca9Add aGoogleDecisionsModelbackend for the Gemini Decisions API (#13181)8d7ed03Accept prose before the closing JSON fence inPR Review Checksanswers (#13...c429ff0[voice-authority] Note text that looks like an application frame in user and ...aefc43aSummarize the tail messagesSummarizingCompaction(keep_user_messages=True)...96fa403Report whenTieredCompactioncan't reachtarget_tokensbecause its tiers ...3dc712aAllow aCancellationTokenfired from a workflow signal to cancel a Temporal...86727dfComplete a Gemini Live turn that ends silently after a tool result (#13126)03fd4e3EnableWebFetchToolin the gh-aw shim only when theANTHROPIC_BASE_URLho...7ae4e2cLower amax_tokensabove a Claude model's maximum output to it instead of s...fda76d3AddOutputTokenLimitExceeded, and fall back to the next model in `ModelFall...- Additional commits viewable in compare view
Updates pytest-asyncio from 1.4.1.post1 to 1.5.0
Release notes
Sourced from pytest-asyncio's releases.
pytest-asyncio v1.5.0
1.5.0 - 2026-11-12
Added
- A
fixture-scopevalue forasyncio_default_fixture_loop_scope. With it, an asynchronous fixture withoutloop_scopeuses itsscopeas its loop scope. This is the behavior of an unset option in pytest-asyncio 1.x. (#1095)Fixed
- Invalid arguments to the
asynciomarker, including an invalidloop_scope, are reported with a message that names the test instead of a traceback. (#1584)
Commits
0c7805echore: Prepare release of v1.5.0.0d50e49Prepare v1.5 (#1610)- See full diff in compare view
Updates ruff from 0.17.4 to 0.17.5
Release notes
Sourced from ruff's releases.
0.17.5
Release Notes
Released on 2026-11-12.
Preview features
- Redirect renamed rules (#30081)
- [
flake8-async] Implementpytest-raises-exception-group(ASYNC401) (#28772)- [
flake8-logging-format] MarkG004fix as unsafe (#30087)- [
pylint] Implementno-method-argument(PLE0211) (#29748)- [
refurb] Do not offer fix whenwriteis nested in an expression (FURB103) (#29789)- [
ruff] Addrenamed-rule(RUF202) (#30079)- [
ruff] Expandredirected-noqato other suppression comments (RUF101) (#30078)Bug fixes
- Avoid crashing when Ruff logs to a closed stderr pipe (#30209)
- Fix fixes that add or remove a keyword argument with a parenthesized value (#30189)
- [
flake8-import-conventions] Avoid shadowing names when fixing import aliases (ICN001) (#30035)- [
flake8-bugbear] FixB010autofix for parenthesizedsetattrcalls (#30168)- [
flake8-pyi] FixPYI034autofix for parenthesizedselfannotations (#30173)- [
pyupgrade] FixUP031duplicating closing parentheses around the format string (#30172)Rule changes
- [
flake8-bugbear] Ignore frozen dataclass instantiations (B008) (#29231)Performance
- Various performance improvements (#30375, #30211, #30368, #30259, #30354, #30115, #30133, #30132, #30358, #30135, #30369, #30351, #30352, #30347, #30379, #30328, #30329, #29446)
Other changes
- Fix FreeBSD source builds with libc 0.2.190 (#30176)
Contributors
... (truncated)
Changelog
Sourced from ruff's changelog.
0.17.5
Released on 2026-11-12.
Preview features
- Redirect renamed rules (#30081)
- [
flake8-async] Implementpytest-raises-exception-group(ASYNC401) (#28772)- [
flake8-logging-format] MarkG004fix as unsafe (#30087)- [
pylint] Implementno-method-argument(PLE0211) (#29748)- [
refurb] Do not offer fix whenwriteis nested in an expression (FURB103) (#29789)- [
ruff] Addrenamed-rule(RUF202) (#30079)- [
ruff] Expandredirected-noqato other suppression comments (RUF101) (#30078)Bug fixes
- Avoid crashing when Ruff logs to a closed stderr pipe (#30209)
- Fix fixes that add or remove a keyword argument with a parenthesized value (#30189)
- [
flake8-import-conventions] Avoid shadowing names when fixing import aliases (ICN001) (#30035)- [
flake8-bugbear] FixB010autofix for parenthesizedsetattrcalls (#30168)- [
flake8-pyi] FixPYI034autofix for parenthesizedselfannotations (#30173)- [
pyupgrade] FixUP031duplicating closing parentheses around the format string (#30172)Rule changes
- [
flake8-bugbear] Ignore frozen dataclass instantiations (B008) (#29231)Performance
- Various performance improvements (#30375, #30211, #30368, #30259, #30354, #30115, #30133, #30132, #30358, #30135, #30369, #30351, #30352, #30347, #30379, #30328, #30329, #29446)
Other changes
- Fix FreeBSD source builds with libc 0.2.190 (#30176)
Contributors
Commits
c91de8aSkip the W605 source scan when a literal has no decoded backslash (#30376)0ebb77dBump version to 0.17.5 (#30380)1236ea2Defer the metaclass check in PYI034 (#30373)0b43597Parse hex escapes one ASCII byte at a time (#30374)427ae55Move thegroupedoutput format intoruff_db(#30290)f96ec54Apply per-file ignores as rule sets (#30375)489b338[ty] Renamecall-top-callabletocall-to-function-with-unknown-signature...3ffff35Speed up streaming notebook reads (#30379)75938f6Short-circuit typing member checks before resolving qualified names (#30369)5a784dfInline publicCursormethods across crates (#30368)- Additional commits viewable in compare view
Updates ty from 0.0.99 to 0.0.102
Release notes
Sourced from ty's releases.
0.0.102
Release Notes
Released on 2026-11-12.
Breaking changes
- Split
invalid-return-typeinto 3 focussed rules; renameempty-bodyand <...Description has been truncated
Summary by cubic
Bumps the
python-minor-patchgroup to bringlangchain-core,langchain,pydantic-ai,pydantic-ai-slim,openai,pytest-asyncio,ruff, andtyup to their latest minor or patch releases.Breaking changes
langchain-core1.8.0 andlangchain1.5.0 drop support for Python 3.10, which reached end-of-life; projects on Python 3.10 will keep resolving the previous versions.ty0.0.102 renamed two rules, so the ignore codes instackone_ai/tools.pywere updated to match (unresolved-attribute→missing-attribute,invalid-mutable-override→incompatible-mutable-override).