Fix/pr 7714 7715 review issues#7764
Merged
Merged
Conversation
…ng-20260517' into fix/pr-7714-7715-review-issues
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…4-7715-review-issues
…er.ts Function was inserted before the os/path imports during the PR 7757 merge; moved it to the server-start section where it is actually called. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…nputs, apt branch init, wire all publish jobs into release-all - windows-store-release.yml: fix submit-to-store if-condition from `inputs.submit_to_store == true` to `!= false` so the string 'true' from workflow_dispatch is handled correctly (GitHub boolean inputs are strings in expressions) - publish-android-update-manifest.yml: upgrade actions/checkout@v4 to @v6 for consistency with the rest of the repo - publish-apt-repo.yml: add orphan branch creation step before the apt-repo checkout so the workflow doesn't fail on first run when the apt-repo branch doesn't yet exist in the remote - elizaos-os-full-release.yml: add prepare job to compute bare semver version (strip leading 'v' from github.ref_name) before calling publish-apt-repo.yml; GitHub Actions has no substring() function so the stripping must happen in a shell step with an output; also adds prepare to publish-apt needs - release-all.yml: replace snap-notify reminder stub with a real snap job that calls snap-publish.yml; add windows-store job calling windows-store-release.yml after desktop succeeds; add android-update-manifest job that computes version_code from semver and dispatches publish-android-update-manifest.yml via gh workflow run (required because reusable workflow calls cannot use computed per-step values as inputs); add aosp-update-manifest job calling publish-aosp-update-manifest.yml after linux-os-iso; add apt-repo job calling publish-apt-repo.yml after linux-os-deb; update summary job needs list and result table to include all new jobs Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Staged all unstaged changes brought in by the swiftun/nubs/messylinux branch merges: cloud-frontend branding and feature additions, UI voice-pill refactor (CSS rename), shared-brand exports, homepage/os-homepage/usb-installer updates, design-assets cleanup, stale script removal, and biome lint fixes. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…b shell, vendor bundler
- Add iOS sideloading via Dadoum/Sideloader: SideloaderIosBackend (ideviceid/ideviceinfo
device detection, Apple ID auth + 2FA via subprocess, IPA download/sign/install pipeline,
EU DMA + Japan SCA regional notice via timezone detection)
- Add IosFlasher.tsx: 8-screen UI (scan → device → region notice → app select → Apple ID
login → 2FA → progress → complete), password never stored post-POST
- Add DependencyManager: auto-detects adb/fastboot/libimobiledevice/sideloader from
~/.elizaos/flasher/vendor/bin/ and PATH; auto-installs via brew/apt/winget with
GitHub binary fallback and manual instructions fallback
- Add DependencyGate.tsx: startup screen with per-dep status icons, auto-install buttons,
collapsible manual instructions, two-click bypass
- Add InstallerShell.tsx: unified 3-tab shell (💾 USB / 📱 Android / 🍎 iPhone)
wrapping existing FlasherApp + new IosFlasher; update main.tsx entry point
- Add vendor/download.mjs: downloads platform-tools + Sideloader binaries to
~/.elizaos/flasher/vendor/bin/{platform}/ at postinstall time; partial-success tolerant
- Add vendor/checksums.json + .gitignore for vendor/bin/
- Add weekly update-vendor-checksums.yml CI workflow
- Fix Samsung T7 (USB-NVMe enclosure) detection in macOS USB installer backend:
add Ejectable field check for enclosures that report Removable:false but Ejectable:true
- Wire /dependencies, /ios/* endpoints into aosp-flasher server.ts
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Remove two aspirational TODO(cloudflare) comments with no current action (app-factory.ts subdomain wiring, app-domains.tsx DNS targets) - Strip @param/@returns JSDoc lines that merely restate the function signature or parameter name across 17 utility/config/auth files in cloud-shared and cloud-frontend - Preserve all comments that explain non-obvious design decisions, migration dependencies, driver quirks, or intentional constraints Net: -117 lines in target packages (cloud-shared, cloud-frontend). All kept NOTEs and TODOs document real constraints or in-flight work. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Removes a try/catch that silently returned 0 on DB failure for credit balance — a financial field where a wrong value is worse than a visible error. The org row is already fetched in the same Promise.all higher up in fetchDashboardDataInternal, so this also eliminates a duplicate DB query. The inline expression `org ? Number(org.credit_balance) : 0` is equivalent to the old code's happy path. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add pg + @types/pg to root devDependencies so packages/scripts/cloud/admin scripts can resolve pg regardless of cwd (packages/scripts has no package.json) - Add biome-ignore for runtime.useModel() call in plugin-mcp/utils/error.ts — Biome misidentifies the elizaOS `use*` method as a React hook due to naming convention Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Fix two Biome formatter violations in packages/app-core that caused lint failure: collapse a multi-line retrieveConnection call that exceeded the line length limit, and reformat a dynamic import expression in auth-store.ts to match Biome's expected style. Both packages/app and packages/app-core typecheck and lint clean. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…ontend
- Replace redundant console.error+toast pairs with toast-only in app-settings,
apps-table, billing-tab, and eliza-chat-interface (file upload, STT, TTS)
- Promote silent console.error-only catches to toast.error in usage-tab,
affiliates-page-client, app-analytics (3 fetches), app-users, and
eliza-chat-interface message load — errors were invisible to users
- Add explanatory comments to two bare catch{} blocks around
invalidateCorsAllowedPorts dynamic imports in dev-server and eliza
- Fix pre-existing biome formatter issue in training-benchmarks.ts
(historyStmt.all call was split across 4 lines, now inline)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…hared - Replace `Promise<any>` return on room-title evaluator handler with `Promise<void>` - Replace `AsyncIterable<any>` in vercel-ai-gateway with `AsyncIterable<TextStreamPart<...>>` - Replace `[] as any[]` in node-autoscaler test with typed `DockerNode[]` - Add explanatory comments to all remaining justified `as unknown as` casts: - Drizzle `.returning()` type limitation (memories, participants, rooms) - Legacy elizaOS evaluator shape vs current Evaluator interface - Private AgentRuntime.initResolver property access - Content index signature incompatibility with plugin metadata - JSONB column types in sensitive-requests repository - Wadis library type mismatch with local WadisClientLike interface - Generic empty array [] cast for typed T extends unknown[] return - Test mock assignment to globalThis.fetch - Worker polyfill assignments to globalThis builtins - Variadic CompatDatabaseMethod requires any[] for contravariance All 178 tests pass; both packages typecheck clean. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…ard pages
Pages that returned <DashboardLoadingState/> from the !ready/!authenticated
short-circuit dropped their <Helmet><title>...</title></Helmet> during the
auth-loading window, which let the homepage <title> ("Eliza Cloud - Launch
Eliza") leak through. Surfaced by an authed dashboard sweep against the
production preview build with VITE_PLAYWRIGHT_TEST_AUTH=true baked in.
Fixed routes:
/dashboard (was leaking)
/dashboard/agents (race-prone)
/dashboard/api-keys "API Keys"
/dashboard/mcps "MCP Servers"
/dashboard/documents "Knowledge"
Also splits admin dashboard routes out of the renderable e2e fixture into
a dedicated "requires admin auth" smoke — they correctly redirect to /login
under the default test-auth user, so asserting them as renderable was
wrong; now we assert the gate is in place instead.
Adds a new tests/e2e/siwe-flow.spec.ts that exercises the real
@stwd/sdk SIWE handshake (GET /auth/nonce → build EIP-4361 message →
sign locally with viem → POST /auth/verify) end-to-end, using an
ephemeral generated EVM key and mocked Steward endpoints. Validates
the bundle posts a real EIP-4361 message with our nonce and that the
signature recovers to our generated address.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
…arning and fix build Next.js 15 detects two bun.lock files (milady root + eliza root) and picks the wrong workspace root, then fails to find pages-manifest.json. Setting outputFileTracingRoot to the examples root anchors resolution correctly. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Remove unnecessary `as unknown as` double-casts where single `as` suffices (sensitive-request-store, phrase-chunked-tts) - Replace branded-type cast in server-utils.test.ts with proper stringToUuid() - Remove redundant undefined assignment before delete in public-link-adapter.test - Replace process.env cast with typed Record<string,unknown> in onboarding test - Call AgentRuntime.getService() directly instead of casting to inline interface - Replace `as any` in server-role-seeding tests with typed `as unknown as IAgentRuntime` and `as unknown as Memory`, removing biome-ignore suppressions - Document necessary remaining casts with explanatory comments (CapacitorSQLite plugin type gap, TtsHandler/EdgeTtsHandler function-type mismatch, module→subset boundary in auth-store and training-benchmarks) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…vels up) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Contributor
|
Important Review skippedAuto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Contributor
|
Claude encountered an error after 0s —— View job I'll analyze this and get back to you. |
Contributor
LifeOps Benchmark —
|
Contributor
LifeOps Benchmark —
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Relates to
Risks
Background
What does this PR do?
What kind of change is this?
Documentation changes needed?
Testing
Where should a reviewer start?
Detailed testing steps