[daytime-em]

What

Updates both dependencies to their latest stable releases:

	Before	After
media3	1.10.1	1.11.0
Mux Data	1.12.0	1.13.2

The Mux Data artifact coordinate is pinned to the media3 minor version, so data-media3-at_1_10 becomes data-media3-at_1_11. Mux Data 1.13.2 exists specifically to add media3 1.11.x support.

One code change was needed

media3 1.11.0 adds HLS Content Steering and Pathway Cloning, which appended a 13th nullable ContentSteeringInfo parameter to the HlsMultivariantPlaylist constructor. That broke one test fixture in MuxHlsDownloadCallbackTests, which now passes null.

No main-source changes: our offline code only ever reads multivariant playlists (CapturingHlsPlaylistParserFactory, and MuxHlsDownloadCallback building StreamKeys) and never constructs them.

Changelog review

I checked both changelogs for anything affecting the offline download / DRM work that just landed in #106. Nothing does:

• media3 1.11.0 — the only download-related change is a fix (resolve HLS variables in media playlists fetched for download). Content Steering is purely additive.
• Mux Data 1.13.x — audiotrackchange events, play/pause fixes for non-ExoPlayer Players, and a generic-bounds tightening on monitorWithMuxData (* -> Player) that compiles clean here.

Testing

• All 27 library unit tests pass, including the offline/DRM suites from Feature: Offline Playback and Offline DRM  #106
• assembleDebug succeeds for all modules, plus the automatedtests instrumentation APK
• Verified the resolved graph is actually androidx.media3:*:1.11.0 and data-media3-at_1_11:1.13.2

Note for release notes

The Mux Data artifact coordinate change is consumer-visible for anyone who depends on Mux Data directly alongside this SDK.

🤖 Generated with Claude Code

---

Note

Low Risk
Dependency version bumps with a single test-only constructor fix; main offline/DRM paths do not construct multivariant playlists.

Overview
Bumps androidx.media3 from 1.10.1 to 1.11.0 and Mux Data from 1.12.0 to 1.13.2, and switches the Mux artifact from data-media3-at_1_10 to data-media3-at_1_11 so it matches the Media3 minor version.

The only source change is in MuxHlsDownloadCallbackTests: the HlsMultivariantPlaylist test fixture now passes contentSteeringInfo = null for Media3 1.11’s new constructor parameter (HLS Content Steering). Production offline/HLS code is unchanged.

^{Reviewed by Cursor Bugbot for commit 9b51cbb. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

🔎 Observe-only. This grade is informational and does not block merging — no approvals are enforced. We are collecting rulings before turning on any policy.

AI Review Classification — review:low

Requirement to merge: no human approval required (AI review only)

Layer	Result
Rules (engine)	score 3.8 (low)
Routing	route-to-llm
Learned calibration	4.3 (medium)
LLM judge	3.4 (low) — Minor dependency bump of media3 and Mux Data with a corresponding test constructor update for a new HLS content-steering parameter; low blast radius and easily reversible, though playback-SDK upgrades can carry subtle behavioral changes that merit smoke testing.

Full audit — which factors fired and why

PR Risk Assessment — LOW (score 3.8/10)

Signals observed:

• production_repo (+0.3) — pull request targets a production repository [stacking decay x0.6 at rank 3]
• gradle_build_chain (+1.0) — core build-system change (gradle/libs.versions.toml)
• dependency_bump (+2.5) — framework dependency change in gradle/libs.versions.toml (/(?i)media3.*\d+.\d+/)

Reductions applied:

• none

Scoring:

• base (inherent): 3.8
• after reductions: 3.8
• binding floor: none
• final: 3.8 → low

⚠️ Low after reductions, not inherently low. Risk signals fired (see above) but reductions brought the score below the medium threshold. Review the suppressed signals before relying on this.

Recommended review: review:low — automated AI review is sufficient; no human required.

Grade this call — react on this comment: 👍 right · 👎 wrong · 🚀 graded too low · 👀 graded too high.
Strongly disagree? Reply /ai-review disagree should=<trivial|low|medium|high> <reason>.