feat(providers): accounts, keys, Custom presets, proxy, usage per account, Codex async questions and reset credits - #1264
Conversation
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub. |
|
Important Review skippedToo many files! This PR contains 418 files, which is 318 over the limit of 100. To get a review, reduce the PR to 100 files or fewer by splitting it into smaller PRs or changing its base branch. Upgrade to a paid plan to raise the limit. Usage-priced reviews support at most 300 files. ⚙️ Run configurationConfiguration used: Repository: arul28/ADE/.coderabbit.yaml Review profile: CHILL Plan: Advanced Run ID: ⛔ Files ignored due to path filters (27)
📒 Files selected for processing (418)
You can disable this status message by setting the 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 |
|
@greptile review |
Bugbot couldn't run - usage limit reachedBugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit. A user or team admin can review and increase usage limits in the Cursor dashboard. (requestId: serverGenReqId_fbde4d5f-24c1-4432-85f8-662fbe815c66) |
|
@CodeRabbit review |
|
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
…teration 2) Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Problem
ADE ran one identity per provider, read from the brain's own environment. A user with two Claude subscriptions, an OpenRouter key, and a ChatGPT plan could not mix them, and every Custom agent-plus-model setup had to be rebuilt by hand. Alongside that, six shipped bugs (YAML 1.1 booleans in slash-command discovery, the whole-tree turn diff on an uncommitted tree, the lost multi-question note, the never-re-arming Codex rate-limit warning, the hardcoded Claude config home, the Droid home variable mismatch), Codex async questions and reset credits were unhandled, and the chat-actions pane kept stale "running" rows for a day after their runtimes died.
Change and boundary
resume_returndialogs render through the question card.--arg/--input-jsonoverrides on every launch surface, includinglanes create-from-linear --start-chat.Deliberately not changed: the model picker stays one entry per model; accounts are chosen only in Settings; no vendor login enters the account vault.
Verification
/quality: five full passes on the branch plus an 18-pass ship revalidation on the rebased tree; gate empty; Windows parity clean on every pass./test: prune clean; proxy test folder consolidated; parity for docs, iOS, CLI, TUI, and SDK; analytics gate met withade_feature_usedat the durable owner boundaries.xcrun swiftc -parse; noxcodebuildran, because free disk is below the 30 GB floor.Authored with Claude Fable 5.1 via ADE, with Claude Opus 5 and GPT-5.6 Luna subagents.
Note
Medium Risk
Changes RPC role semantics, removes orchestration service from the CLI runtime, and alters how CLI sessions resolve credentials and presets at launch—areas that affect agent tooling and session resume behavior.
Overview
Removes the orchestrator RPC identity and drops desktop
orchestrationServicewiring from the CLI brain (including orchestrator event emission and chat’sgetOrchestrationService). ADE RPC now treats coordination callers asagent, always advertises resources on initialize, and tightens who gets local computer-use tools accordingly; docs drop the orchestration service doc-map entry and trim personal-chat create notes tied to orchestrator-lead mode.Extends
start_cli_sessionwithinstanceId,presetId, andcredentialId, resolving Claude/Codex accounts and harness presets at launch time and persisting those ids in PTY resume metadata (including preset id when resolution is gated/unavailable). Runtime bootstrap gains lazyproxyService, optionalruntimeSocketPathfor chat ownership, proxy hooks in sync, and small analytics wiring (pending-input dismiss, usage → internal analytics).Hardens CLI launch planning across
ade new,ade chat create,shell start-cli, andlanes create-from-linear --start-chat: shared validation for merged--arg/--input-jsonoverrides (provider, permissions, spawn lineage, instance/preset/credential), plus newade proxyandade providers accountscommand surfaces with formatters/tests. README/doc tweaks and a largecli.test/adeRpcServer.testrefresh mirror the role rename and new behaviors.Reviewed by Cursor Bugbot for commit 477a825. Configure here.