Redesign usage share card around model activity

[Chipagosfinest]

Status

Ready for maintainer review. Merge-risk follow-up cf1157cd preserves the established summary card as the default and exposes the redesigned model-activity card as an explicit preview choice. The underlying activity-card implementation retains its exact-head local release-gate proof from ccc4f585970899129a13efe6af1c4fd8a8d381d9; GitHub CI validates the current PR head and merge ref.

What changed

• preserves the existing summary card as the default while offering the model-activity card through an explicit Summary / Model activity selector
• keeps preview, copied PNG, saved PNG, and copied text on the same selected card style
• makes token volume and dated daily activity the primary story of the opt-in model-activity card
• preserves model route identity as source instance + provider + raw model before public sanitization
• keeps provider/source and model independently readable (Claude Sonnet 4 via OpenRouter)
• carries selected window bounds, partial daily coverage, unavailable totals, pricing coverage, hidden-route counts, and native currencies into the share payload
• uses a separate token-complete model projection so routes with known tokens survive unavailable pricing without changing the cost-complete dashboard projection
• renders locally without prompts, account labels, emails, identifiers, paths, projects, or raw events
• exports exact 1200×630, 600×315, and 300×158 PNGs from the final SwiftUI renderer

Real renderer proof

These are deterministic, redacted outputs produced by ShareStatsRenderer from the committed proof fixture on the exact PR head—not design concepts.

1200×630

Additional social sizes

• 600×315 renderer output
• 300×158 renderer output
• Accessible text export

Renderer assertions verify exact pixel dimensions, nonblank color diversity, and fully opaque sampled output at all three sizes.

Renderer provenance

This PR preserves the final no-Metal renderer merged in #2292. The merged tree at 6ba3d195 uses NSHostingView + NSBitmapImageRep + NSGraphicsContext + displayIgnoringOpacity; it does not use SwiftUI ImageRenderer. The maintainer’s merge verification explicitly records that ImageRenderer/Metal was removed for Intel CI compatibility.

Against current main, this PR only parameterizes the bitmap pixel dimensions for deterministic social-size exports and updates the default filename. It does not replace the proven rendering path.

Truth-contract fixes

• Selected dates: periodEnd comes from the selected chart domain’s inclusive end, so quiet trailing days stay in a 30-day snapshot.
• Partial activity: missing, invalid, or overflowed daily totals remain unavailable; they are never converted to confirmed zero.
• Route identity: typed model-row IDs prevent delimiter collisions, and the payload keeps source-instance identity through ranking and sanitization.
• Hidden routes: observed, shareable, shown, and hidden routes are counted separately; overflow copy cannot claim all routes are shown after sanitization.
• Spend: native currencies remain separate, known subtotals survive missing pricing, and partial totals use lower-bound language.
• Token/pricing independence: token-known routes remain visible when their price is unavailable; cost-only rows do not enter token rankings.
• Privacy: account display labels are not exported. Same-provider instances receive public numbered labels, and private model suffixes are truncated.

Validation

Activity-card implementation commit: ccc4f585970899129a13efe6af1c4fd8a8d381d9

• make test — 718 selections, 60/60 groups passed
• first-pass failed groups — 0
• retries / recovered groups / timeouts — 0 / 0 / 0
• make check — passed; SwiftFormat 0 files, SwiftLint 0 violations across 1,576 files
• swift test --filter ShareStatsTests — 21/21 passed
• proof fixture matrix — 1, 4, 8, and 20 same-model source instances; partial 7/30-day history; unavailable/overflow totals; unavailable pricing; mixed native currencies; private labels/suffixes; selected quiet trailing days; exact renderer sizes
• production builder regression — an OpenRouter CostUsageTokenSnapshot with known Claude tokens and unavailable pricing survives SpendDashboardModel.build and appears as Claude Sonnet 4 via OpenRouter

Compatibility follow-up commit: cf1157cd

• restores the established summary renderer as ShareStatsCardStyle.defaultStyle
• adds an explicit Summary / Model activity preview selector
• threads the selected style through preview, PNG copy/save, and text copy
• adds regression coverage proving default output equals Summary and differs from Model activity
• local Swift parse and SwiftFormat passed; full local Swift Testing/SwiftLint require full Xcode frameworks unavailable in the active Command Line Tools installation, so current-head GitHub CI is the authoritative gate

Proof file SHA-256:

38398a27ae0d0e1cbe227c19f86a2c8b819434bec8de1637fb651ac69c6bb93c  share-stats-1200x630.png
ba2dab49223ae49da61e32c96f66aae7e0c72279099b4e3e52902671dd5cb33a  share-stats-600x315.png
914a23c4ceb2d6d4837467a56663979f9bc0df11c53e728320961f0fc3705670  share-stats-300x158.png
8a19d9a66f5bcb67c7813560c3d0b3c5a2f52fb562c9758898d00a42fe682af8  share-stats.txt

Scope and honest limitations

The committed multi-provider proof fixture demonstrates the renderer and payload contract under multiple provider/source/model routes. The production builder regression demonstrates that a real dashboard input shape preserves an OpenRouter route when pricing is missing.

This PR does not claim every CodexBar provider currently supplies token/cost history to Usage & Spend. The selected share card reads whatever supported sources the existing Usage & Spend pipeline supplies; provider fetcher expansion remains separate work. No real account data was used in the committed proof.

Related work

• Fix black PNG exports from usage share cards #2292 — reliable non-black PNG rendering, merged and included in this rebase
• Fix partial model history aggregation #2308 / Keep complete model rows visible when another source is incomplete #2287 — partial model-history contract, merged and included
• Feature discussion: model-centric token / cost view across providers #2257 — broader model-centric identity discussion
• Separate the scalable usage dashboard from the social share artifact #2290 — bounded dashboard/share-artifact contract
• Track Usage & Spend production hardening and sharing follow-ups #2291 — Usage & Spend hardening tracker
• Add native macOS sharing for aggregate usage cards #2289 — native macOS sharing follow-up

[clawsweeper]

Codex review: needs real behavior proof before merge. Reviewed July 20, 2026, 8:03 PM ET / July 21, 2026, 00:03 UTC.

Summary
The branch retains the summary share card as default and adds a selectable model-activity card with expanded source-aware token, coverage, privacy, and social-image export handling.

Reproducibility: not applicable. This PR proposes an additional user-facing card style, not a report of broken existing behavior. The final interactive selector behavior can be validated through a fresh packaged-app run rather than a bug reproduction.

Review metrics: 2 noteworthy metrics.

• Patch surface: 13 files affected; 2,097 additions, 440 deletions. The feature reaches rendering, payload construction, dashboard modeling, preview/export wiring, tests, and committed proof artifacts rather than being a presentation-only tweak.
• Current-head UI follow-up: 1 selector/default commit after renderer proof. The latest compatibility commit changes interactive preview/export behavior, so prior renderer artifacts do not fully demonstrate the final branch.

Merge readiness
Overall: 🦐 gold shrimp
Proof: 🦐 gold shrimp
Patch quality: 🐚 platinum hermit
Result: blocked until stronger real behavior proof is added.

Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch.

Rank-up moves:

• Run the freshly built CodexBar.app at cf1157cd and capture redacted evidence for Summary and Model activity selections.
• Show that Copy Image, Copy Text, and Save export the style currently selected in the preview; redact any private account or usage data.
• After updating the PR body with that evidence, request a fresh review if one is not triggered automatically.

Proof guidance:

• [P1] Needs stronger real behavior proof before merge: The branch has strong deterministic renderer artifacts for the earlier activity-card head, but the current selector/default follow-up needs redacted fresh-bundle evidence demonstrating both styles and selected-style copy/save/text behavior. After adding proof, update the PR body; ClawSweeper should re-review automatically. If it does not, the PR author or someone with repository write access can comment @clawsweeper re-review.

Risk before merge

• [P1] The new Summary / Model activity selector is a user-facing product surface; preserving the default avoids an upgrade break, but maintainers still need to accept the additional supported narrative and export mode.
• [P1] The current head lacks fresh bundle-level proof that switching styles keeps preview, copied PNG, saved PNG, and copied text synchronized in the final app build.

Maintainer options:

1. Decide the mitigation before merge
   Keep Summary as the default, merge the activity card only with owner approval, and attach redacted current-head bundle proof showing both styles plus Copy Image, Copy Text, and Save producing the selected style.
2. Pause or close
   Do not merge this PR until maintainers decide whether the risk is worth taking.

Next step before merge

• [P1] No narrow automated repair is appropriate: the next step is owner acceptance of the additional share-card mode, followed by contributor-provided final-head runtime evidence.

Maintainer decision needed

• Question: Should CodexBar support the new opt-in Model activity share-card style alongside the established Summary card?
• Rationale: The implementation now avoids replacing the existing default, but accepting a second long-lived share-card narrative and selector is a product-surface decision that source inspection cannot make.
• Likely owner: steipete — The owner has supplied the visual/product context and recent merged work for this sharing and Usage & Spend area.
• Options:
  • Approve the opt-in activity card (recommended): Accept the additional style after current-head bundle proof confirms the selector and all export actions stay synchronized.
  • Keep the summary-only contract: Decline the additional share-card mode and retain the current summary artifact as the sole supported workflow.

Security
Cleared: The diff modifies local Swift rendering, payload, preview, tests, and static proof assets without adding dependencies, network behavior, permissions, workflow changes, or new secret-handling paths.

Review details

Best possible solution:

Keep Summary as the default, merge the activity card only with owner approval, and attach redacted current-head bundle proof showing both styles plus Copy Image, Copy Text, and Save producing the selected style.

Do we have a high-confidence way to reproduce the issue?

Not applicable: this PR proposes an additional user-facing card style, not a report of broken existing behavior. The final interactive selector behavior can be validated through a fresh packaged-app run rather than a bug reproduction.

Is this the best way to solve the issue?

Unclear: retaining Summary as the default is the narrow compatibility-safe implementation, but whether Model activity should be a supported selectable artifact requires owner approval rather than further autonomous code changes.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against f8636cb37eb0.

Label changes

Label justifications:

• P3: This is an optional share-card capability with no demonstrated core-runtime regression after preserving the existing default.
• rating: 🦐 gold shrimp: Overall readiness is 🦐 gold shrimp; proof is 🦐 gold shrimp and patch quality is 🐚 platinum hermit.
• status: 📣 needs proof: The PR needs real behavior proof before ClawSweeper can clear the contributor ask. Needs stronger real behavior proof before merge: The branch has strong deterministic renderer artifacts for the earlier activity-card head, but the current selector/default follow-up needs redacted fresh-bundle evidence demonstrating both styles and selected-style copy/save/text behavior. After adding proof, update the PR body; ClawSweeper should re-review automatically. If it does not, the PR author or someone with repository write access can comment @clawsweeper re-review.
• proof: 📸 screenshot: Contributor real behavior proof includes screenshot evidence. The branch has strong deterministic renderer artifacts for the earlier activity-card head, but the current selector/default follow-up needs redacted fresh-bundle evidence demonstrating both styles and selected-style copy/save/text behavior.

Evidence reviewed

What I checked:

• Default compatibility preserved: The final branch introduces ShareStatsCardStyle.defaultStyle = .summary, retaining the existing card as the default while making Model activity an explicit alternative. (Sources/CodexBar/ShareStatsRenderer.swift:6, cf1157cd40d8)
• Selection is threaded through user actions: The preview controller changes Copy Image, Copy Text, and Save callbacks to receive the chosen card style, so the new selector affects exported artifacts rather than only the preview. (Sources/CodexBar/ShareStatsWindowController.swift:56, cf1157cd40d8)
• Prior renderer and data-contract blockers were addressed: The review discussion documents fixes for quiet-window dates, unavailable daily totals, typed source-aware model identity, and hidden-route accounting in the earlier exact-head implementation; the previous completed review cycle found no remaining code findings on the merged-data version. (Tests/CodexBarTests/ShareStatsTests.swift:1, ccc4f5859708)
• Current-head proof gap: The supplied deterministic renderer artifacts and full local validation are tied to ccc4f585; the latest cf1157cd selector follow-up reports parse/format checks but not a freshly built bundle run proving both selected styles and all export actions. (Sources/CodexBar/ShareStatsWindowController.swift:56, cf1157cd40d8)
• Repository UI policy applied: The repository policy requires freshly built-bundle validation when UI/runtime behavior changes and prefers stable seams for menu behavior; this PR changes the preview UI and export wiring. (AGENTS.md:1, f8636cb37eb0)
• Product tracking relationship: The open dashboard/share-artifact tracker explicitly identifies this PR as the draft implementation for the social share-card contract, while the broader model-centric discussion remains open for design direction. (f8636cb37eb0)

Likely related people:

• steipete: The owner supplied the original before/after product context, authored the merged partial-model-history follow-up, and is the most direct decision-maker for the supported share-card narrative. (role: product-direction owner and recent area contributor; confidence: high; commits: 6ba3d19575fd, bddbc9aff14e; files: Sources/CodexBar/ShareStatsRenderer.swift, Sources/CodexBar/SpendDashboardModel.swift)
• Chipagosfinest: Earlier merged sharing and Usage & Spend work establishes feature-history context beyond this proposed branch, including the existing local share workflow that this PR extends. (role: original sharing-workflow contributor; confidence: medium; commits: 7b11885ba51f, 0397529ae6e5; files: Sources/CodexBar/ShareStatsCardView.swift, Sources/CodexBar/ShareStatsPayload.swift)

What the crustacean ranks mean

• 🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
• 🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
• 🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
• 🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
• 🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
• 🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
• 🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

How this review workflow works

• ClawSweeper keeps one durable marker-backed review comment per issue or PR.
• Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
• A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
• PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
• Maintainers can also comment @clawsweeper review to request a fresh review only.
• Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
• Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
• Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

Review history (8 earlier review cycles)

• reviewed 2026-07-18T19:18:56.175Z sha a77e8aa :: needs real behavior proof before merge. :: none
• reviewed 2026-07-18T19:42:08.537Z sha a77e8aa :: needs real behavior proof before merge. :: [P1] Keep the selected window end when activity is quiet
• reviewed 2026-07-18T21:59:16.922Z sha a77e8aa :: needs real behavior proof before merge. :: [P1] Preserve the selected window end | [P1] Represent unavailable daily token history | [P2] Stop calling aggregated rows source pairs | [P2] Count hidden routes before overflow copy
• reviewed 2026-07-18T22:44:25.522Z sha 4df2ed3 :: needs real behavior proof before merge. :: [P1] Rebase onto the current PNG renderer
• reviewed 2026-07-18T23:43:08.902Z sha ccc4f58 :: needs maintainer review before merge. :: none
• reviewed 2026-07-19T05:54:35.216Z sha 16adf9c :: needs maintainer review before merge. :: none
• reviewed 2026-07-19T17:33:06.451Z sha be59f07 :: needs maintainer review before merge. :: none
• reviewed 2026-07-19T18:36:22.404Z sha 79f7560 :: needs maintainer review before merge. :: none

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a77e8aafd5

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".

[steipete]

Before (current card, synthetic data)

Fresh 1200×630 render from current origin/main (618a6826e06de9974f7d5fdab7ef770e5162c4c6) using the existing ShareStatsTests synthetic dashboard fixture.

After (PR design target — deterministic concept)

The PR's committed 1200×630 design-target image at a77e8aafd59c22e016891683799bf7fed3ce085f.

The fresh BEFORE and attempted runtime AFTER used the same existing ShareStatsTests synthetic dashboard fixture. The committed design target is a concept image rather than a renderer export and contains its own illustrative values.

Runtime-render note: on the exact PR head, swift test --jobs 4 --filter ShareStatsTests passed all 11 tests, but exporting that same synthetic fixture through ShareStatsRenderer produced a black 1200×630 PNG on visual inspection. I therefore did not present it as a real rendered AFTER. No real account data was used.

[Chipagosfinest]

@clawsweeper re-review

Exact-head proof handoff for ccc4f585970899129a13efe6af1c4fd8a8d381d9:

Real renderer outputs

All three files were generated by the final ShareStatsRenderer from the committed redacted fixture. They are not design concepts.

• 1200×630 PNG — SHA-256 38398a27ae0d0e1cbe227c19f86a2c8b819434bec8de1637fb651ac69c6bb93c
• 600×315 PNG — SHA-256 ba2dab49223ae49da61e32c96f66aae7e0c72279099b4e3e52902671dd5cb33a
• 300×158 PNG — SHA-256 914a23c4ceb2d6d4837467a56663979f9bc0df11c53e728320961f0fc3705670
• Accessible text export — SHA-256 8a19d9a66f5bcb67c7813560c3d0b3c5a2f52fb562c9758898d00a42fe682af8

Renderer tests assert exact pixel dimensions, PNG signature, nonblank sampled color diversity, and sampled alpha > 0.99 at every size.

Findings addressed

• selected quiet trailing days use the selected inclusive chart-domain end
• unavailable/partial/overflow daily history remains unavailable and uses lower-bound language, never confirmed zero
• typed model route identity preserves source instance + provider + raw model before public sanitization
• observed/shareable/hidden/shown route counts are distinct
• token-known routes survive unavailable pricing without changing the cost-complete dashboard projection
• native currency subtotals and partial pricing coverage remain explicit
• private account labels, identifiers, paths, prompts, and unsafe model suffixes are excluded

Each of the four original inline threads has a finding-specific reply and is resolved.

Exact-head gates

• make test: 718 selections, 60/60 first-pass groups, 0 failed groups, 0 retries, 0 recovered groups, 0 timeouts; 663.4 seconds
• make check: passed; SwiftFormat 0 files; SwiftLint 0 violations across 1,576 files
• swift test --filter ShareStatsTests: 21/21 passed
• fixture matrix: 1/4/8/20 source instances; same model across sources; partial 7/30 coverage; unavailable/overflow totals; unavailable pricing; mixed native currencies; private labels/suffixes; quiet final days; all three renderer sizes
• production builder path: OpenRouter input with known Claude tokens and unavailable pricing remains visible as Claude Sonnet 4 via OpenRouter

The branch was rebased and exact-head tested on main at 618a6826e06de9974f7d5fdab7ef770e5162c4c6, including merged #2292 and #2308. During the full gate, main advanced only through unrelated StepFun-only #2312; GitHub merge-ref CI covers the newer base.

Scope note: the multi-provider fixture proves the renderer/payload contract, and the OpenRouter test proves the production dashboard-builder path. This PR does not claim every CodexBar provider already supplies usage history to Usage & Spend; provider fetcher expansion is separate.

[clawsweeper]

🦞🧹
ClawSweeper re-review requested.

I asked ClawSweeper to review this item again.
Action: item re-review queued (workflow sweep.yml, event repository_dispatch).
Result: the existing ClawSweeper review comment will be edited in place when the review finishes.

Re-review progress:

• State: Complete
• Detail: The durable review comment was published and its verdict was queued for target-serialized routing.
• Run: https://github.com/openclaw/clawsweeper/actions/runs/29665717417
• Updated: 2026-07-18T23:47:15.093Z

[Chipagosfinest]

CI triage update for ClawSweeper/maintainers:

• Linux arm64 release build succeeded.
• Its test step failed only in the known concurrent PlatformGatingTests invocation-log race: https://github.com/steipete/CodexBar/actions/runs/29664394226/job/88132288185
• Exact assertion: PlatformGatingTests.swift:128, Claude CLI runtime skips logged out interactive fallback; both .auto and .cli parameter cases started concurrently.
• This PR changes no Linux platform-gating or Claude CLI code.
• The narrow test-only containment is Serialize flaky Claude CLI platform-gating cases #2311, which is non-draft, ClawSweeper-rated 🦞 diamond lobster, and already has green Linux x64 + arm64 proof.
• Cross-reference added to Serialize flaky Claude CLI platform-gating cases #2311 with this fresh independent reproduction.

The share-card exact-head local gate remains 718 selections / 60 of 60 groups with no retries or timeouts. GitHub lint, GitGuardian, and the arm64 release build are green; remaining CI continues.

[Chipagosfinest]

Renderer provenance correction for ClawSweeper

The P1 in the generated review is based on a reversed premise. This PR does not replace a proven ImageRenderer path.

Immutable repository evidence:

• Merged prerequisite Fix black PNG exports from usage share cards #2292 at 6ba3d195 contains NSHostingView + NSBitmapImageRep + NSGraphicsContext + displayIgnoringOpacity: merged source.
• The final no-Metal implementation was maintainer-authored in c53e6516. ImageRenderer was deliberately removed for Intel CI compatibility.
• The maintainer’s merge verification explicitly confirms that decision: “renderer uses NSHostingView + NSBitmapImageRep + NSGraphicsContext and does not reintroduce ImageRenderer/Metal”: merge record.
• Against current upstream/main, this PR retains that exact renderer and only adds deterministic pixelSize output dimensions plus the usage-oriented filename: current head.

Exact-head runtime proof at ccc4f585970899129a13efe6af1c4fd8a8d381d9:

• 1200×630 PNG — SHA-256 38398a27ae0d0e1cbe227c19f86a2c8b819434bec8de1637fb651ac69c6bb93c
• 600×315 PNG — SHA-256 ba2dab49223ae49da61e32c96f66aae7e0c72279099b4e3e52902671dd5cb33a
• 300×158 PNG — SHA-256 914a23c4ceb2d6d4837467a56663979f9bc0df11c53e728320961f0fc3705670
• Renderer tests decode each exact size and reject uniform/blank output.
• swift test --filter ShareStatsTests: 21/21 passed.
• make check: passed, 0 format/lint violations.
• make test: 718 selections, 60/60 groups, 0 failures, 0 retries, 0 timeouts.
• git diff --check: clean.

The only red CI job is the unrelated, freshly reproduced parameterized PlatformGatingTests concurrency flake, isolated in #2311 and cross-referenced here. Please reevaluate the renderer finding against the final merged tree of #2292, not its superseded initial ImageRenderer proposal.

[Chipagosfinest]

@clawsweeper re-review

[Chipagosfinest]

@steipete this is the remaining product-layer follow-through for the share-card work that has already landed: #2112 introduced sharing, #2116 added Usage & Spend, #2292 fixed the production PNG renderer, and #2308 established the partial-history truth contract. This PR brings those pieces together into the model-activity share card; leaving it open means the recently merged prerequisites are present, but the intended public artifact is not.

I have now merged current main (f8636cb37) into the branch at 79f756022, with no conflicts or overlapping code changes. On the combined stack: make check is green; focused share-card tests are 21/21; dashboard-model tests are 28/28; and the full local suite completed 720 selections / 60 groups with 0 failures, retries, or timeouts. Current exact-head CI: https://github.com/steipete/CodexBar/actions/runs/29697669492

The implementation/proof is ready; the remaining decision is the product call ClawSweeper identified: whether model activity and explicit partial-coverage disclosure should be the default share narrative. I would appreciate that owner decision so this final piece does not get stranded after its prerequisites merged.

@clawsweeper re-review

[clawsweeper]

🦞🧹
ClawSweeper re-review requested.

I asked ClawSweeper to review this item again.
Action: item re-review queued (workflow sweep.yml, event repository_dispatch).
Result: the existing ClawSweeper review comment will be edited in place when the review finishes.

Re-review progress:

• State: Complete
• Detail: The durable review comment was published and its verdict was queued for target-serialized routing.
• Run: https://github.com/openclaw/clawsweeper/actions/runs/29699007984
• Updated: 2026-07-19T18:39:39.245Z