chore(deps-dev): bump pydantic-ai-slim from 2.49.0 to 2.53.0

[dependabot]

Bumps pydantic-ai-slim from 2.49.0 to 2.53.0.

Release notes

Sourced from pydantic-ai-slim's releases.

v2.53.0 (2026-10-01)

🛡️ Security

This release fixes one security issue in ConcurrencyLimitedModel. See the advisory for full details and affected versions.

• GHSA-6fqq-452j-qhrp (high): a streamed request through ConcurrencyLimitedModel or limit_model_concurrency could keep its concurrency slot when the slot was released on a different task than the one that acquired it: after an early exit (the consumer stopped iterating, raised, or was cancelled), and also after fully consuming stream_text() with its default debouncing. Repeated streams could then block every request sharing the limiter. Agent-level max_concurrency and non-streaming requests are not affected. Reported by @​lche511. (#9478)

The fix also changes how limiters are shared: a model wrapper now raises UserError when it shares a limiter with the agent making the request or with an enclosing model wrapper, ConcurrencyLimiter.acquire() takes a slot on every call, even on the same task, and a custom AbstractConcurrencyLimiter must allow release() from another task.

Patched in 2.53.0. v1 is not affected.

What's Changed

⚠️ Compatibility Notes

• Add oneOf schema support to TestModel's generated data by @​pydanty in pydantic/pydantic-ai#8783
• Deprecate pydantic_ai_harness.step_persistence.naming by @​mpfaffenberger in pydantic/pydantic-ai#9499

🚀 Features

• Count Google grounding web search queries in usage by @​pydanty in pydantic/pydantic-ai#8891
• Add AbsurdDurability to the harness as a replacement for pydantic-ai-absurd by @​adtyavrdhn in pydantic/pydantic-ai#8946
• Let AskUser defer questions to the host and time out a slow answerer by @​mpfaffenberger in pydantic/pydantic-ai#9509
• Add SystemOneModel to run decision models such as CLM and Laya over the /v1/systemone API by @​mpfaffenberger in pydantic/pydantic-ai#8942
• Add ToolCallJudge to assess tool calls before execution by @​DouweM in pydantic/pydantic-ai#9041
• Send images and documents from GPT-Live tool results to the delegated backend by @​DouweM in pydantic/pydantic-ai#9048
• Expose the message-history repair pipeline as repair_messages by @​DouweM in pydantic/pydantic-ai#8370

🐛 Bug Fixes

• Kill the process group when a LocalWorkspaceBackend.run timeout fires during process startup by @​dsfaccini in pydantic/pydantic-ai#9358
• Avoid invalid-escape warning floods during CodeMode analysis by @​mpfaffenberger in pydantic/pydantic-ai#9475
• Keep the registered Memory toolset across runs so durable execution accepts it by @​mpfaffenberger in pydantic/pydantic-ai#9489
• Keep the registered Planning toolset across runs so durable execution accepts it by @​mpfaffenberger in pydantic/pydantic-ai#9507
• Keep planted files from escaping BubblewrapSandbox on the next launch by @​dsfaccini in pydantic/pydantic-ai#9457
• Render non-finite eval metrics as inf/-inf/nan by @​pydanty in pydantic/pydantic-ai#8852
• Snap gemini-3.1-flash-image thinking efforts to minimal and high on the Gemini API by @​dsfaccini in pydantic/pydantic-ai#9515
• Raise UnexpectedModelBehavior instead of running the wrong tool when function tool calls share a tool_call_id by @​pydanty in pydantic/pydantic-ai#8782
• Build one streamed XaiModel thinking part per output, matching the non-streamed response by @​DouweM in pydantic/pydantic-ai#9416
• Leave image generation calls out of the OpenAIResponsesModel replay when openai_store=False by @​DouweM in pydantic/pydantic-ai#9388
• Build the same ModelResponse from streamed and complete OpenAIChatModel and OpenRouterModel responses by @​DouweM in pydantic/pydantic-ai#9418
• Make OpenAIResponsesModel build the same ModelResponse from a stream as from a complete response by @​DouweM in pydantic/pydantic-ai#9417
• Make xAI push-to-talk reply only when asked: commit_audio() alone no longer triggers a reply, and create_response() is always answered by @​DouweM in pydantic/pydantic-ai#9070
• Tolerate OpenAI realtime status, status_details and session values the SDK does not know yet by @​DouweM in pydantic/pydantic-ai#9392
• Stop SubAgents from forcing thinking on disk agents by @​DouweM in pydantic/pydantic-ai#9384
• Restore DynamicWorkflow reveal announcements after compaction or history loss by @​DouweM in pydantic/pydantic-ai#9371
• Make Logfire Temporal spans replay-safe by @​DouweM in pydantic/pydantic-ai#7006
• Allow unschemable types in the derived Temporal ActivityConfig schema by @​pydanty in pydantic/pydantic-ai#9577
• Let GoogleRealtimeModel declare tools whose dict values are recursive models by @​dsfaccini in pydantic/pydantic-ai#9607

New Contributors

• @​Poojita060926 made their first contribution in pydantic/pydantic-ai#8900

Full Changelog: pydantic/pydantic-ai@v2.52.0...v2.53.0

... (truncated)

Commits

• b38ab5d Document that InlineDefsJsonSchemaTransformer only walks object and array k...
• c50e223 Keep the test suite passing on anthropic 1.11.0, dbos 3.2.0 and `google-g...
• 453f19f Fix RuntimeError and a leaked slot when streaming through `ConcurrencyLimit...
• d63315d Let GoogleRealtimeModel declare tools whose dict values are recursive mod...
• ce6b675 Allow unschemable types in the derived Temporal ActivityConfig schema (#9577)
• f596e4b Keep the launch-directory workspace in clai2 when capability functions supp...
• 5b5a414 Fix the TOML disk sub-agent test that still expected the removed effort floor...
• 1013241 Make Logfire Temporal spans replay-safe (#7006)
• 9da37d4 Expose the message-history repair pipeline as repair_messages (#8370)
• aba3e75 Say native tools aren't routed to OpenAI realtime yet, rather than unsupporte...
• Additional commits viewable in compare view

[github-actions]

Codecov Results 📊

✅ 133451 passed | ⏭️ 7276 skipped | Total: 140727 | Pass Rate: 94.83% | Execution Time: 450m 29s

📊 Comparison with Base Branch

Metric	Change
Total Tests	📈 +3059
Passed Tests	📈 +3061
Failed Tests	📉 -2
Skipped Tests	—

➖ Removed Tests (1)

View removed tests

• test_binary_content_encoding_mixed_content[False-False]
  • File: tests.integrations.litellm.test_litellm

All tests are passing successfully.

✅ Patch coverage is 100.00%. Project has 2592 uncovered lines.
❌ Project coverage is 90.26%. Comparing base (f5bf69b) to head (d036581).

Coverage diff

@@            Coverage Diff             @@
##        master       #PR       +/-##
==========================================
- Coverage    90.28%    90.26%    -0.02%
==========================================
  Files          202       202         —
  Lines        26625     26625         —
  Branches      9926      9926         —
==========================================
+ Hits         24039     24033        -6
- Misses        2586      2592        +6
- Partials      1515      1514        -1

---

Generated by Codecov Action

[alexander-alderman-webb]

Resolved by 1ca0326

[dependabot]

OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting @dependabot ignore this major version or @dependabot ignore this minor version.

If you change your mind, just re-open this PR and I'll resolve any conflicts on it.