Repository navigation
fix(fork): resolve upstream merge 3d78e915e3 (2026-11-06) - #86
Merged
Merged
Conversation
- AntigravityAdapterV2.test.ts: keeps the fork's flavorOptions object, which a later test spreads, and takes upstream's `defaultModel: Effect.undefined`. The old suppression line goes with `Effect.succeed(undefined)`. - Orchestrator.ts: mayStartSubagentFollowup keeps the fork's delegatedCompletion check (the port of upstream pingdotgg#28462) and takes upstream's new return, which also starts usage-limit continuations. - ws.ts: keeps the fork's no-auth branch, which skips the revocation race, and runs upstream's new framedRpcWebSocketHttpEffect in both arms. - settings.ts: both sides only added keys; keeps the fork's previewBrowserMode and upstream's speech settings, in ServerSettings and in ServerSettingsPatch. - provider-acp adapter.ts: keeps the fork's runtimeAccount and upstream's openAgentRequests at runtime spawn and at runtime replacement. - pnpm-lock.yaml: regenerated from upstream's file with --lockfile-only; it is identical to the textual merge. The fork's code also follows upstream's new Effect rules and APIs: - EnvironmentAuth.ts and one fork test read the environment through HostProcess.Environment; PreviewDisplayError is a Schema.TaggedError. - Fork test fakes gain upstream's new methods: the fake Claude sessions, the AssetSigner mock and the secret store's label. - pdfWorker.ts returns pdf.js's own loading task with destroy wrapped. The worker still stops five minutes after the last document closes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Thread transfer impact✅ Thread transfer remains within every enforced ceiling.
Baseline: unavailable · PR result: Scenario and decoded snapshot size10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.
Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed. |
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
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.
This pull request resolves the merge of upstream 3d78e91 (159 commits) into the fork and updates the fork's own code to match upstream's changes. The fork still differs from upstream in the same 182 files as before, and adds no new lint suppression.
Upstream edited code next to five of the fork's own changes, so the merge left conflicts. Each conflict keeps both the fork's addition and upstream's. The base branch holds git's merge with its conflict markers, so this pull request shows only the hand-made part.
Upstream also made three code-style rules into errors, so the fork's code now reads the environment and defines one error the way upstream does. Upstream added methods to some interfaces that fork tests fake, so those fakes gain them. Upstream's new PDF test needs the PDF worker to return the library's own loading task. The worker still stops five minutes after the last PDF closes.
The routing of Claude usage reads, reset-credit reads and redeems is unchanged. Selected tests and typechecks of the changed packages pass. The full check of files changed against upstream also passes (evidence).
Files
Conflicts:
AntigravityAdapterV2.test.ts: the fork's sharedflavorOptionsand upstream'sdefaultModel.Orchestrator.ts: the fork's check that a finished delegated task may start a follow-up for its parent beside upstream's new usage-limit clause.ws.ts: sessions without auth still skip the sign-out race; both paths use upstream'sframedRpcWebSocketHttpEffect.packages/contracts/src/settings.ts: the fork'spreviewBrowserModeand upstream's speech settings.packages/provider-acp/src/server/adapter.ts: the fork'sruntimeAccountand upstream'sopenAgentRequests.pnpm-lock.yamlis not a conflict: upstream's file pluspnpm install --lockfile-onlyequals git's merge, so this pull request leaves it unchanged.Rule and interface updates:
EnvironmentAuth.tsand one fork test read throughHostProcess.Environment;PreviewDisplayErroris aSchema.TaggedError; the fake Claude sessions and the fake secret resolver gain upstream's new members; the tools test provides an emptyAssetSignermock, which upstream's tool handlers now need;pdfWorker.tsreturns pdf.js's loading task withdestroywrapped.