[github-actions]

Automated weekly dependency update via python scripts/update-matrix-latest.py && uv lock --upgrade.

⚠️ Provider SDK packages changed. A human needs to re-record cassettes locally before merging.

Summary by Sourcery

Upgrade the Python dependency matrix and lockfile to the latest supported package releases.

Enhancements:

• Update the dependency matrix to newer releases across provider SDKs, agent frameworks, evaluation tools, and supporting libraries.

Build:

• Regenerate the Python lockfile with upgraded dependency resolutions.

Tests:

• Refresh dependency versions used by the Python integration test matrix; provider cassette recordings require re-recording for the updated SDKs.

[sourcery-ai]

Reviewer's Guide

This automated weekly update advances the matrix’s latest dependency versions and regenerates the uv lockfile, while retaining historical version entries for compatibility testing. Because several provider SDKs changed, reviewers should verify the lockfile resolution and require local cassette re-recording before merge.

Flow diagram for the weekly dependency matrix update

flowchart LR
    A[Run update-matrix-latest.py] --> B[Update latest provider versions]
    B --> C[Retain historical matrix versions]
    C --> D[Run uv lock --upgrade]
    D --> E[Regenerate uv.lock]
    E --> F[Re-record provider cassettes locally]

Loading

File-Level Changes

Change	Details	Files
Refresh the dependency matrix’s latest provider and tooling versions while preserving the existing compatibility pins.	Upgrade latest entries for provider SDKs and integrations, including OpenAI, Anthropic, Cohere, Agents, LiteLLM, Google, Pydantic AI, and related frameworks. Update supporting packages such as Temporal, pytest, and other matrix dependencies. Regenerate the uv lockfile to resolve the upgraded dependency graph; the large lockfile diff is not individually reviewed here.	py/pyproject.toml py/uv.lock

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[sourcery-ai]

Hey - I've found 1 issue

Fixed security issues:

• anyio (link)
• certifi (link)
• cryptography (link)
• gitpython (link)
• mcp (link)
• pyjwt (link)
• python-multipart (link)
• starlette (link)
• urllib3 (link)
• uv (link)

Prompt for AI Agents

Please address the comments from this code review:

## Individual Comments

### Comment 1
<location path="py/pyproject.toml" line_range="287" />
<code_context>

 [tool.braintrust.matrix.openai]
-latest = "openai==2.33.0"
+latest = "openai==3.42.1"
 "1.92.0" = "openai==1.92.0"
 "1.77.0" = "openai==1.77.0"
</code_context>
<issue_to_address>
**Latest integration replays fail**

When CI runs the updated latest-version integration sessions before matching cassettes are re-recorded, `_run_tests` selects `cassettes/latest/`, and VCR's `record_mode="none"` cannot record updated interactions; cassette mismatches fail the tests or leave them exercising stale behavior.

Re-record and commit the `latest` cassettes for the updated SDK interactions.

Also at `py/pyproject.toml:293`, `py/pyproject.toml:297`, `py/pyproject.toml:301`, `py/pyproject.toml:306`, `py/pyproject.toml:310`, `py/pyproject.toml:314`, `py/pyproject.toml:319`, `py/pyproject.toml:331`, `py/pyproject.toml:335`, `py/pyproject.toml:339-340`, `py/pyproject.toml:344`, `py/pyproject.toml:348`, `py/pyproject.toml:352`, `py/pyproject.toml:359`, `py/pyproject.toml:363`, `py/pyproject.toml:367`, `py/pyproject.toml:371`, `py/pyproject.toml:375`, `py/pyproject.toml:379`.
</issue_to_address>

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

[sourcery-ai]

🟠 High · Latest integration replays fail

When CI runs the updated latest-version integration sessions before matching cassettes are re-recorded, _run_tests selects cassettes/latest/, and VCR's record_mode="none" cannot record updated interactions; cassette mismatches fail the tests or leave them exercising stale behavior.

Re-record and commit the latest cassettes for the updated SDK interactions.

Also at py/pyproject.toml:293, py/pyproject.toml:297, py/pyproject.toml:301, py/pyproject.toml:306, py/pyproject.toml:310, py/pyproject.toml:314, py/pyproject.toml:319, py/pyproject.toml:331, py/pyproject.toml:335, py/pyproject.toml:339-340, py/pyproject.toml:344, py/pyproject.toml:348, py/pyproject.toml:352, py/pyproject.toml:359, py/pyproject.toml:363, py/pyproject.toml:367, py/pyproject.toml:371, py/pyproject.toml:375, py/pyproject.toml:379.

Prompt for AI agents

In `py/pyproject.toml` at line 287:

**Latest integration replays fail**

When CI runs the updated latest-version integration sessions before matching cassettes are re-recorded, `_run_tests` selects `cassettes/latest/`, and VCR's `record_mode="none"` cannot record updated interactions; cassette mismatches fail the tests or leave them exercising stale behavior.

Re-record and commit the `latest` cassettes for the updated SDK interactions.

Also at `py/pyproject.toml:293`, `py/pyproject.toml:297`, `py/pyproject.toml:301`, `py/pyproject.toml:306`, `py/pyproject.toml:310`, `py/pyproject.toml:314`, `py/pyproject.toml:319`, `py/pyproject.toml:331`, `py/pyproject.toml:335`, `py/pyproject.toml:339-340`, `py/pyproject.toml:344`, `py/pyproject.toml:348`, `py/pyproject.toml:352`, `py/pyproject.toml:359`, `py/pyproject.toml:363`, `py/pyproject.toml:367`, `py/pyproject.toml:371`, `py/pyproject.toml:375`, `py/pyproject.toml:379`.

From AGENTS.md L30-34:

6. **Prefer real integration coverage over mocks.**
   - For provider/integration behavior, prefer VCR-backed tests with checked-in cassettes.
   - This includes bugs in tracing/span shaping that happen after the SDK returns a real provider payload. If the behavior depends on the provider's actual response shape, treat it as VCR-first work, not mock-first work.
   - Be actively skeptical of mock/fake tests for provider integrations. Do not reach for mocks just because they are faster or easier to write.
   - Avoid mocks/fakes unless the code is purely local or there is no practical cassette-based option.