Restore z.ai rate-window pace - #2802
Merged
steipete merged 4 commits intoAug 10, 2026
Merged
Conversation
clawsweeper
Bot
added
rating: 🦪 silver shellfish
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 894a6ee431
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
clawsweeper
Bot
added
P2
|
Codex review: needs changes before merge. Reviewed August 9, 2026, 8:21 PM ET / August 10, 2026, 00:21 UTC. ClawSweeper reviewWhat this changesThe PR restores z.ai time-limit durations, adds tightly scoped pace rules for five-hour, weekly, and MCP-monthly windows, and expands regression coverage. Regression provenancePossible regression — probable (reproduction; reviewed change). No predecessor PR is attributed. Merge readinessThis PR fixes the verified five-hour and weekly z.ai pace path, but it does not actually render the claimed MCP-monthly pace in the menu because the extra-window renderer still rejects z.ai. A small code-and-test follow-up is needed before merge. Priority: P2 Review scores
Verification
How this fits togetherThe z.ai plugin converts quota API limits into normalized primary, secondary, and MCP extra usage windows. Descriptor pace rules feed CLI output and menu-card rendering, where the MCP window follows the extra-window display path. flowchart LR
A[z.ai quota API] --> B[JavaScript adapter]
B --> C[Primary, weekly, and MCP windows]
C --> D[z.ai pace descriptor]
D --> E[CLI pace output]
C --> F[Menu extra-window renderer]
F --> G[Menu card pace detail]
Before merge
Findings
Agent review detailsSecurityNone. Review metrics
Root-cause clusterRelationship: Members:
Proposal only: this assessment does not dispatch repair, suppress jobs, mutate sibling items, close, or merge anything. Merge-risk optionsMaintainer options:
Technical reviewBest possible solution: Route the z.ai MCP extra window through descriptor-backed reset-window pace rendering, then add a menu-card regression that proves the monthly pace appears while rolling 30-day token and credit windows remain excluded. Do we have a high-confidence way to reproduce the issue? Yes, from source: a z.ai MCP extra window reaches the menu extra-window renderer, whose current provider gate excludes z.ai before its descriptor pace rule can run. Is this the best way to solve the issue? No; the adapter and descriptor repair is sound for five-hour and weekly windows, but it must also connect the MCP extra-window menu path to reset-window pace rendering. Full review comments:
Overall correctness: patch is incorrect AGENTS.md: found and applied where relevant. Codex review notes: model internal, reasoning high; reviewed against 648a948b7ab4. LabelsLabel justifications:
EvidenceAcceptance criteria:
What I checked:
Likely related people:
Rank-up movesOptional improvements that raise the rating; they are not merge blockers.
Rating scale
Overall follows the weaker of proof and patch quality. Workflow
HistoryReview history (3 earlier review cycles)
|
clawsweeper
Bot
removed
the
merge-risk: 🚨 compatibility
steipete
force-pushed
the
kiran/restore-zai-mcp-monthly-pace-20260809
branch
from
August 10, 2026 00:15
3ab4720 to
a6f15b9
Compare
clawsweeper
Bot
added
proof: sufficient
Owner
|
Post-merge proof:
Scope caveat: this PR fixes the remaining z.ai regression only. #2431 tracks multiple providers and remains open. |
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
2 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.
Addresses the remaining z.ai regression in #2431.
The JavaScript provider migration kept computing
TIME_LIMITdurations but stopped exporting them. That dropped the established monthly MCP marker and left the z.ai descriptor without its verified 5-hour and weekly pace lanes.This restores the one-minute MCP marker as the existing calendar-month sentinel, preserves explicit
TIME_LIMITdurations, leaves unknown cadence unset, and enables pace only for the verified 5-hour and weekly slots. Monthly inference also requires the adapter's MCP marker, so an explicit rolling 30-day token or credit window retains its duration without being rewritten as a calendar month.Verification:
CI=1 CODEXBAR_SUPPRESS_TEST_KEYCHAIN_ACCESS=1 swift test --disable-keychain --disable-netrc --filter "ZaiProviderTests|ProviderPaceCapabilityTests|CLISnapshotTests|ProviderPresentationPolicyCharacterizationTests|UsagePaceTextTests|ZaiMenuCardTests|MenuBarPaceTextTests"— 98 tests passedCI=1 CODEXBAR_SUPPRESS_TEST_KEYCHAIN_ACCESS=1 ./Scripts/test-plugin-engines.sh— 76 tests passed under QuickJS and 76 under JavaScriptCoremake check— passedCODEXBAR_SUPPRESS_TEST_KEYCHAIN_ACCESS=1 make test— all 833 selections passed in 70 groups; no retries or timeoutsgit diff --check origin/main...HEAD— passedswift build -c release --product CodexBarCLI— passedReal provider proof
A maintainer-owned z.ai QA credential was injected at runtime through the approved 1Password service-account path; the key value, account identity, and usage percentages were never printed or added to the PR. The same command was run before and after the patch.
api; real primary window5-hour;windowMinutes: 300; reset present; no pace object.api; the same5-hour/ 300-minute/reset-bearing window;pace.primarypresent; no provider error.This directly exercises the reported real-provider regression while confirming that authentication and window selection remain unchanged.
Compatibility risk is low: token and credit parsing is unchanged, explicit rolling cadences are preserved, and unknown MCP cadence remains fail-closed.
Review focus: the one-minute marker mapping, the MCP-only monthly discriminator, and the exact z.ai descriptor lane rules.