[moq-bot]

🤖 New release

• rind: 0.2.0 -> 0.2.1
• rind-jsonwebtoken: 0.2.0 -> 11.0.0
• rind-rustls: 0.2.0 -> 0.23.0
• rind-dimpl: 0.2.0 -> 0.9.0
• rind-rcgen: 0.1.14 -> 0.14.0
• rind-object-store: 0.2.0 -> 0.14.0
• rind-ring: 0.2.0 -> 0.17.0

Changelog

rind

0.2.1 - 2026-11-07

Added

• [breaking] each adapter's version matches the upstream it adapts (#440)
• (bench) measure Apple M4 natively in the bench report (#435)
• (cpu) hold DIT across a batch of calls, putting M1 AES-GCM ahead of aws-lc-rs (#431)
• (aes) AES-192 in every AES mode on every backend (#414)
• (digest) MD5 under digest::legacy (#412)

Fixed

• (ctgrind) type the armv7 key-wrap-with-padding test's lengths (#426)

Other

• (release) ship rind as 0.2.1, not 0.3.0 (#457)
• (aarch64) run s2n-bignum's faster _alt P-256 and P-384 Montgomery multiplication (#434)
• (aes-gcm) seal under 129 bytes on by-8 on the AVX tier (#436)
• (aes-gcm) pad Vale's partial blocks in registers, and re-set the open crossovers (#430)
• test iOS in the simulator and Android under cross (#432)
• (aes-gcm) check 16-byte tags as whole words, so by-8 opens no longer trail its seals (#429)
• (ed25519) verify without ADX through curve25519-dalek's AVX2 backend (#415)
• (aarch64) run s2n-bignum's by-8 AES-128-GCM kernels on cores with SHA3 (#408)
• (aes-gcm) hash and cipher partial blocks without memcpy on x86_64, and seal short messages on by-8 on VAES-AVX2 (#409)
• (ecdsa) derive a key pair's public key on its first read (P-256 generation ahead of BoringSSL) (#407)

rind-jsonwebtoken

11.0.0 - 2026-11-07

Added

• (adapters) [breaking] rind-jsonwebtoken and rind-instant-acme check the CPU in default_provider() (#445)
• [breaking] each adapter's version matches the upstream it adapts (#440)

Other

• (release) ship rind as 0.2.1, not 0.3.0 (#457)

rind-rustls

0.23.0 - 2026-11-07

Added

• [breaking] each adapter's version matches the upstream it adapts (#440)

Other

• (release) ship rind as 0.2.1, not 0.3.0 (#457)
• test iOS in the simulator and Android under cross (#432)

rind-dimpl

0.9.0 - 2026-11-07

Added

• [breaking] each adapter's version matches the upstream it adapts (#440)

Other

• (release) ship rind as 0.2.1, not 0.3.0 (#457)

rind-rcgen

0.14.0 - 2026-11-07

Added

• (rind-rcgen) key_identifier() fills in key identifiers as rcgen does (#452)
• (rind-rcgen) check() and serial_number() so users need no direct rind dependency (#451)
• [breaking] each adapter's version matches the upstream it adapts (#440)

Other

• (release) ship rind as 0.2.1, not 0.3.0 (#457)

rind-object-store

0.14.0 - 2026-11-07

Added

• [breaking] each adapter's version matches the upstream it adapts (#440)

Other

• (release) ship rind as 0.2.1, not 0.3.0 (#457)

rind-ring

0.17.0 - 2026-11-07

Added

• (rind-ring) patch ring from moq-dev/rind-patch, without rind's test vectors (#456)
• a ring patch shim replaces ring across a binary's tree (#444)
• [breaking] each adapter's version matches the upstream it adapts (#440)

Other

• (release) ship rind as 0.2.1, not 0.3.0 (#457)
• test iOS in the simulator and Android under cross (#432)

---

This PR was generated with release-plz.

[pullfrog]

Your OpenCode Go account is out of credit. Top it up, then re-trigger Pullfrog.

OpenCode Go's response

provider error: Insufficient account funds

Top up OpenCode Go →

  ^{｜ Rerun failed job ➔ ｜ View workflow run ｜ via Pullfrog ｜ Using glm-5.3}

[kixelated]

Automated review by review (OpenAI)

Reviewed head bff415dd18b4b7fd0cfb2d7fbb70c6ef29df5518 against base feddff13c013b62d83ecdf4559aac9f815def86b.

No actionable findings. The current delta only updates uuid 1.29.0 → 1.29.1 in Cargo.lock. The checksum matches the registry index, and the unchanged Rust 1.89 minimum fits this workspace's 1.90 requirement. Both locked hax-lib-macros-types versions enable UUID v4/default features; the upstream patch leaves the v4/RNG implementation unchanged. The timestamp/v7 fixes and unstable zerocopy gating do not change that consumer path.

Direction: the scoped dependency update is reasonable; no code changes requested. The release-version/changelog changes listed in the description are already in the base and are not part of this delta.

Verification: GitHub-only static review of the lockfile, dependency metadata, and upstream diff. No local build or tests run. Queue/build checks are currently skipped, so this is not a passing-build claim.

(Written by review, OpenAI)

[kixelated]

Automated review by review (OpenAI)

Incremental review of 2b07034cd8bdc40a4f7baefcda3af961a0f35871 against base dc2cc33e5dc1172e6bc2bf78a97a8a1dc27c4e57, following the review of bff415dd18b4b7fd0cfb2d7fbb70c6ef29df5518.

No new actionable findings. All seven publishable crate versions match their changelogs and lock entries, versioned rind dependencies advance consistently to 0.2.1, and rind-instant-acme remains unpublished. The #407 release note matches the merged change since 0.2.0. The UUID update is unchanged from the previous review. Direction: release metadata is consistent; no changes requested.

Verification: GitHub-only static review of the complete 16-file delta, workspace manifests, release configuration, and lockfile comparison. Exact-head checks are skipped, including the intentionally deferred merge-queue build; this is not a passing-build claim. No local commands, tests, packaging or publication dry run performed.

(Written by review, OpenAI)

[kixelated]

Automated review by review (OpenAI)

Incremental review of dcd05f9b2c74db2e40e20a104f44d51f4f8c8de2 against base 202777dabe14d21723a96cf0e638d98541b887fc, following reviewed head 2b07034cd8bdc40a4f7baefcda3af961a0f35871.

No new actionable findings. After accounting for the base advances, the new delta adds the #412 legacy MD5 and #409 AES-GCM release notes; both match changes already merged into the base. The other 15 file patches are unchanged apart from hunk positions, including crate/dependency versions and the previously reviewed UUID update. Direction: the release metadata remains consistent; no changes requested.

Verification: GitHub-only static comparison of current and previously reviewed deltas, changelog, merged history, and release configuration. Exact-head checks are skipped, including the deferred merge-queue build. No local commands, tests, proofs, packaging, or publication dry run performed; this is not a passing-build claim.

(Written by review, OpenAI)