Skip to content

fix(ui): scope usage requests by agent filter - #87222

Merged
steipete merged 6 commits into
openclaw:mainfrom
Alix-007:issue-87132-usage-agent-filter
May 29, 2026
Merged

fix(ui): scope usage requests by agent filter#87222
steipete merged 6 commits into
openclaw:mainfrom
Alix-007:issue-87132-usage-agent-filter

Conversation

@Alix-007

@Alix-007 Alix-007 commented May 27, 2026 •
edited by steipete
Loading

Copy link
Copy Markdown
Contributor

Fixes #87132.

Summary

  • replace query-token agent scoping with an explicit Usage agent scope selector
  • request all configured agents by default with agentScope: "all", while selected agents still use agentId
  • merge all-agent usage.cost summaries across configured agents and keep scoped cache entries separate
  • keep legacy gateway fallbacks for agentId and agentScope, with generated protocol/docs updates

Verification

  • node scripts/run-vitest.mjs ui/src/ui/controllers/usage.node.test.ts ui/src/ui/app-render-usage-tab.test.ts ui/src/ui/views/usage.test.ts --reporter=dot (18 tests passed)
  • node scripts/run-vitest.mjs run --config test/vitest/vitest.gateway-methods.config.ts src/gateway/server-methods/usage.test.ts src/gateway/server-methods/usage.cost-usage-cache.test.ts src/gateway/server-methods/usage.sessions-usage.test.ts --reporter=dot (34 tests passed)
  • pnpm check:test-types
  • pnpm protocol:check
  • node scripts/run-oxlint.mjs --tsconfig config/tsconfig/oxlint.core.json src/gateway/protocol/schema/sessions.ts src/gateway/server-methods/usage.ts src/gateway/server-methods/usage.test.ts src/gateway/server-methods/usage.cost-usage-cache.test.ts src/gateway/server-methods/usage.sessions-usage.test.ts ui/src/ui/controllers/usage.ts ui/src/ui/controllers/usage.node.test.ts ui/src/ui/app.ts ui/src/ui/app-view-state.ts ui/src/ui/app-render-usage-tab.ts ui/src/ui/app-render-usage-tab.test.ts ui/src/ui/views/usage.ts ui/src/ui/views/usage.test.ts ui/src/ui/views/usageTypes.ts
  • git diff --check
  • .agents/skills/autoreview/scripts/autoreview --mode commit --commit d67156a3c5 --thinking low --no-web-search

Real behavior proof

Behavior addressed: The Usage page needed a real agent scope, not a query-string side effect. Default Usage now asks the gateway for all configured agents, selected agents use agentId, and the gateway can aggregate all-agent cost totals instead of silently falling back to the default agent.

Real environment tested: Local OpenClaw checkout on branch pr-87222, using the Control UI usage tests, gateway usage method tests, generated gateway protocol check, and test typecheck.

Exact steps or command run after this patch: Ran the verification commands listed above after the final fixup commit.

Evidence after fix: UI usage tests passed 18/18, gateway usage tests passed 34/34, pnpm check:test-types passed, pnpm protocol:check passed with generated Swift protocol output staged, targeted oxlint returned success, git diff --check returned clean, and the narrowed autoreview reported no accepted/actionable findings on the Swift compatibility fix.

Observed result after fix: With no selected agent, the UI sends agentScope: "all" to sessions.usage and usage.cost. With a selected agent, both requests send that agentId. The gateway lists sessions across configured agents for explicit all-agent scope, rejects conflicting all-agent plus specific-agent/key requests, aggregates cost totals by date across configured agents, and keeps cost cache entries separated by all/default/specific-agent scope.

What was not tested: I did not run a live browser against a multi-agent gateway. A local pnpm build attempt was stopped after tsdown stayed quiet for about 15 minutes; CI will cover the full build lane. A Testbox pnpm check:changed run exited 255 without actionable logs.

@openclaw-barnacle openclaw-barnacle Bot added app: web-ui App: web-ui size: S proof: supplied External PR includes structured after-fix real behavior proof. labels May 27, 2026
@clawsweeper

clawsweeper Bot commented May 27, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Codex review: needs changes before merge. Reviewed May 28, 2026, 8:57 PM ET / 00:57 UTC.

Summary
The PR adds an explicit Usage agent selector/all-agent scope, threads agentId/agentScope through UI and gateway usage requests, updates cost aggregation/cache behavior, and refreshes protocol docs plus generated Swift models.

PR surface: Source +277, Tests +484, Docs +2, Other +4. Total +767 across 17 files.

Reproducibility: yes. from source: current main's UI never sends agentId, while current main's gateway resolves omitted agentId to the default agent for sessions.usage. I did not run a live multi-agent browser/gateway repro in this read-only review.

Review metrics: 2 noteworthy metrics.

  • Default usage scope: 1 default behavior changed. The empty Usage selector now requests all configured agents instead of the shipped default-agent scope, which is compatibility-sensitive.
  • Gateway protocol surface: 1 request field added, generated Swift model updated. Adding agentScope to sessions.usage changes the operator-client wire contract and requires aligned docs/generated clients before merge.

Merge readiness
Overall: 🦐 gold shrimp
Proof: 🦞 diamond lobster
Patch quality: 🦐 gold shrimp
Result: needs maintainer review before merge.

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

Rank-up moves:

  • [P2] Scope selected-agent fallback aggregates/daily chart data and add a focused view regression.
  • Keep maintainer acceptance explicit for the all-agent default and operator.read visibility change.

Mantis proof suggestion
A real Control UI multi-agent smoke would materially prove the visible selector and gateway-scoped Usage behavior beyond unit tests. A maintainer can ask Mantis to capture proof by posting a new PR comment that starts with the OpenClaw Mantis account mention, followed by:

visual task: verify the Usage page all-agent default and selecting a non-main agent scopes sessions, totals, and cost charts without showing other agents.

Risk before merge

  • [P1] The empty Usage selector changes from shipped default-agent behavior to an explicit all-configured-agents request, so existing operators may see broader usage data after upgrade.
  • [P1] The new agentScope: "all" path lets the existing operator.read usage methods return all configured agents' usage, which is probably intentional here but is still a security-boundary decision.
  • [P2] Selected-agent legacy fallback still needs aggregate and daily-chart scoping so older gateways do not show mixed-agent summaries after dropping unsupported agent parameters.

Maintainer options:

  1. Fix selected-agent fallback scoping (recommended)
    Scope aggregate panels and daily chart data to the selected agent when legacy gateway fallback returns broader session/cost data, with a focused view regression test.
  2. Accept the all-agent default explicitly
    Maintain the empty selector as all configured agents only if maintainers are comfortable changing the upgrade default and operator.read visibility boundary.
  3. Pause for a narrower canonical path
    If the all-agent default is not intended, pause this branch and land a smaller default-agent-plus-selector fix instead of changing the default scope.
Copy recommended automerge instruction 
@clawsweeper automerge

Special instructions:
Update the Usage view so selected-agent mode uses the agent-scoped session set for aggregate panels and filters or derives daily chart/cost data consistently when legacy fallback returns multi-agent results; add a focused regression in ui/src/ui/views/usage.test.ts covering selected agent with multi-agent sessions/costDaily after fallback. Keep the explicit all-agent gateway protocol behavior unchanged.

Next step before merge

  • [P2] A narrow automated repair can fix the selected-agent fallback aggregate/daily scoping; maintainers still own the all-agent default and security-boundary acceptance.

Security
Needs attention: No supply-chain concern found, but the diff deliberately broadens usage visibility under existing operator.read methods and needs maintainer acceptance.

Review findings

  • [P2] Scope selected-agent fallback aggregates — ui/src/ui/views/usage.ts:275-291
Review details

Best possible solution:

Land one canonical fix that scopes selected-agent fallback data consistently, keeps the protocol/docs/generated clients aligned, and has maintainer acceptance for the all-agent default before closing the linked issue and superseding sibling PRs.

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

Yes, from source: current main's UI never sends agentId, while current main's gateway resolves omitted agentId to the default agent for sessions.usage. I did not run a live multi-agent browser/gateway repro in this read-only review.

Is this the best way to solve the issue?

No, not quite yet: the main current-gateway path is a maintainable fix, but selected-agent legacy fallback still needs aggregates and daily chart data scoped to the selected agent. The all-agent default also needs maintainer acceptance as an intentional compatibility/security-boundary change.

Full review comments:

  • [P2] Scope selected-agent fallback aggregates — ui/src/ui/views/usage.ts:275-291
    When a legacy gateway rejects the new agent params, loadUsage retries without server-side agent scoping. In that path the view filters visible rows/totals with agentScopedSessions, but aggregate panels still fall through to sortedSessions and the daily chart still uses data.costDaily, so selecting one agent can display provider/model/tool or daily cost data from other agents. Please scope those derived datasets to the selected agent and add a focused view regression for the fallback shape.
    Confidence: 0.83

Overall correctness: patch is incorrect
Overall confidence: 0.84

AGENTS.md: found and applied where relevant.

Codex review notes: model gpt-5.5, reasoning high; reviewed against 8bd4736f03bc.

Label changes

Label justifications:

  • P2: This is a normal-priority regression fix for multi-agent Usage visibility with limited blast radius but real operator impact.
  • merge-risk: 🚨 compatibility: The PR changes default Usage results on upgrade from default-agent only to all configured agents and adds legacy fallback behavior.
  • merge-risk: 🚨 security-boundary: The PR intentionally broadens usage visibility under existing operator.read methods through explicit all-agent scope.
  • rating: 🦐 gold shrimp: Overall readiness is 🦐 gold shrimp; proof is 🦞 diamond lobster and patch quality is 🦐 gold shrimp.
  • status: ⏳ waiting on author: ClawSweeper has contributor-facing work open and is waiting for author action. Override: A maintainer applied proof: override for this PR.
Evidence reviewed

PR surface:

Source +277, Tests +484, Docs +2, Other +4. Total +767 across 17 files.

View PR surface stats
Area Files Added Removed Net
Source 8 311 34 +277
Tests 6 489 5 +484
Docs 1 3 1 +2
Config 0 0 0 0
Generated 0 0 0 0
Other 2 5 1 +4
Total 17 808 41 +767

Security concerns:

  • [medium] Confirm all-agent usage read boundary — src/gateway/server-methods/usage.ts:989
    The new all-agent path makes sessions.usage bypass the default-agent scope when agentScope: "all" is sent, while the method remains under operator.read; this is likely intended for the Usage page but should be accepted as a boundary change before merge.
    Confidence: 0.78

Acceptance criteria:

  • [P1] node scripts/run-vitest.mjs ui/src/ui/views/usage.test.ts ui/src/ui/app-render-usage-tab.test.ts ui/src/ui/controllers/usage.node.test.ts --reporter=dot.
  • [P1] git diff --check.

What I checked:

  • Root policy read: Read the full root AGENTS.md and scoped guides for UI, gateway, gateway protocol, server methods, docs, and scripts; protocol and compatibility guidance affected the merge-risk review. (AGENTS.md:1, 8bd4736f03bc)
  • Current-main UI source reproduces the missing request scope: On current main, loadUsage sends sessions.usage and usage.cost without any agentId or all-agent scope, so the UI cannot request a non-default agent from the existing controller path. (ui/src/ui/controllers/usage.ts:227, 8bd4736f03bc)
  • Current-main gateway defaults list usage to the default agent: On current main, sessions.usage resolves omitted agentId through resolveDefaultAgentId(config), matching the linked regression report that non-main usage disappears from the Web UI. (src/gateway/server-methods/usage.ts:877, 8bd4736f03bc)
  • PR head adds explicit all-agent scope: At PR head, the UI sends agentScope: "all" by default and agentId when selected, while the gateway adds agentScope handling for sessions.usage and all-agent cost aggregation. (ui/src/ui/controllers/usage.ts:295, d67156a3c552)
  • Remaining fallback-scoping defect: At PR head, selected-agent totals use agentScopedSessions, but aggregate sessions still fall through to sortedSessions and daily chart data stays data.costDaily, so legacy fallback can mix other agents into selected-agent panels. (ui/src/ui/views/usage.ts:275, d67156a3c552)
  • Regression provenance: The default-agent scoping behavior came from the merged gateway session lookup scoping PR, which is the root cause identified by the linked issue. (src/gateway/server-methods/usage.ts:877, 6a12c6f79915)

Likely related people:

  • pgondhi987: PR fix(gateway): scope session data lookups by agent [AI] #81386 and merge commit 6a12c6f added the default-agent scoping that made the Web UI omission visible. (role: introduced current scoped behavior; confidence: high; commits: 6a12c6f79915; files: src/gateway/server-methods/usage.ts, src/gateway/protocol/schema/sessions.ts)
  • steipete: Assigned this PR, applied the proof override, and authored the latest PR commits adding explicit all-agent usage scope and generated Swift compatibility. (role: recent follow-up owner; confidence: high; commits: 9932a64b65da, d67156a3c552; files: src/gateway/server-methods/usage.ts, ui/src/ui/controllers/usage.ts, apps/shared/OpenClawKit/Sources/OpenClawProtocol/GatewayModels.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.

@clawsweeper clawsweeper Bot added rating: 🦪 silver shellfish Thin PR readiness signal; proof, validation, or implementation needs work. status: 📣 needs proof The PR needs real behavior proof before ClawSweeper can clear the contributor ask. P2 Normal backlog priority with limited blast radius. merge-risk: 🚨 compatibility 🚨 May break existing users, config, migrations, defaults, or upgrade paths. labels May 27, 2026
@clawsweeper

clawsweeper Bot commented May 27, 2026

Copy link
Copy Markdown
Contributor

ClawSweeper PR egg

🎁 Pass real behavior proof to wake the egg and unlock a hatchable treat.

Where did the egg go?
  • The egg game starts only after the PR passes the real-behavior proof check.
  • Before that, no creature or rarity is rolled. The treat waits for real proof.
  • This is still just collectible flavor: proof affects review readiness, not creature quality.

@Alix-007

Alix-007 commented May 27, 2026

Copy link
Copy Markdown
Contributor Author

@clawsweeper re-review

Current head 0a4a020cb32f64246b85b150a192cb3b38098cc2 addresses the compatibility concern from the previous review by falling back when legacy sessions.usage rejects agentId and remembering that compatibility result. The PR body includes current-head proof output, and the current CI/proof checks are passing.

@clawsweeper

clawsweeper Bot commented May 27, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

🦞🧹
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:

@Alix-007

Alix-007 commented May 27, 2026

Copy link
Copy Markdown
Contributor Author

@clawsweeper re-review

Updated head 9e9dc4ab4d2028a1befe3480cf8a7bc25cd5f302 addresses the latest review concern by sending the same resolved agentId to both sessions.usage and usage.cost, with fallback retry/remember behavior removing agentId from both requests for legacy gateways. The PR body now includes current-head focused proof for the scoped sessions+cost payload and legacy fallback tests.

@clawsweeper

clawsweeper Bot commented May 27, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

🦞🧹
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:

@openclaw-barnacle openclaw-barnacle Bot added the gateway Gateway runtime label May 27, 2026
@Alix-007

Alix-007 commented May 27, 2026

Copy link
Copy Markdown
Contributor Author

@clawsweeper re-review

Updated head 1fb24cb8a63a8267364a380eab907ec034bc6551 addresses the gateway-side cost scope finding by threading agentId through the usage.cost handler, scoped cost cache key, and loadCostUsageSummaryFromCache. The PR body includes current-head UI and gateway proof for scoped sessions+cost requests, legacy fallback, and cost cache separation.

@clawsweeper

clawsweeper Bot commented May 27, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

🦞🧹
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:

@clawsweeper clawsweeper Bot added rating: 🧂 unranked krab Not merge-ready due to missing proof or serious correctness/safety concerns. and removed rating: 🦪 silver shellfish Thin PR readiness signal; proof, validation, or implementation needs work. labels May 27, 2026
@BingqingLyu

This comment was marked as spam.

Sign in to view
@Alix-007

Alix-007 commented May 28, 2026

Copy link
Copy Markdown
Contributor Author

Maintainer review/proof request:

Current head 1fb24cb8a63a8267364a380eab907ec034bc6551 has all current PR checks passing, including Real behavior proof. The remaining ClawSweeper ask appears to be live multi-agent Control UI/browser + gateway proof for the non-main agent usage path.

I do not have a real multi-agent browser/gateway setup available to produce stronger live proof from my side. If a maintainer can run or record that verification, or is comfortable accepting the focused UI controller/rendering + gateway handler/cache coverage already in this PR, could you review this PR or apply a proof override? I will avoid adding more commits unless there is a concrete code or proof issue to address.

@steipete steipete self-assigned this May 28, 2026
@openclaw-barnacle openclaw-barnacle Bot added docs Improvements or additions to documentation size: L triage: mock-only-proof Candidate: PR proof only shows tests, mocks, snapshots, lint, typecheck, or CI. and removed size: M proof: supplied External PR includes structured after-fix real behavior proof. labels May 28, 2026
@steipete steipete added the proof: override Maintainer override for the external PR real behavior proof gate. label May 28, 2026
@openclaw-barnacle openclaw-barnacle Bot removed the triage: mock-only-proof Candidate: PR proof only shows tests, mocks, snapshots, lint, typecheck, or CI. label May 28, 2026
@steipete steipete removed the status: ⏳ waiting on author ClawSweeper has contributor-facing work open and is waiting for author action. label May 28, 2026
@clawsweeper clawsweeper Bot added rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: ⏳ waiting on author ClawSweeper has contributor-facing work open and is waiting for author action. rating: 🦐 gold shrimp Decent PR readiness signal, but merge confidence is limited. and removed rating: 🦐 gold shrimp Decent PR readiness signal, but merge confidence is limited. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. labels May 28, 2026
@steipete
steipete force-pushed the issue-87132-usage-agent-filter branch 2 times, most recently from 4bea777 to 900637a Compare May 29, 2026 00:10
@steipete
steipete force-pushed the issue-87132-usage-agent-filter branch from 900637a to aacdb12 Compare May 29, 2026 00:30
@steipete
steipete force-pushed the issue-87132-usage-agent-filter branch from aacdb12 to d67156a Compare May 29, 2026 00:51
@steipete steipete removed the status: ⏳ waiting on author ClawSweeper has contributor-facing work open and is waiting for author action. label May 29, 2026
@steipete

steipete commented May 29, 2026

Copy link
Copy Markdown
Contributor

Ready to land.

Verification:

  • gh pr checks 87222 --repo openclaw/openclaw --live --json name,state,link returned no non-green entries at head d67156a3c552c4f9c8b6edf8516b6242bf5cdd26.
  • gh pr view 87222 --repo openclaw/openclaw --json mergeStateStatus,mergeable reported CLEAN / MERGEABLE.
  • Local proof already recorded in the PR body: focused UI usage tests, gateway usage tests, pnpm check:test-types, pnpm protocol:check, targeted oxlint, git diff --check, and autoreview.

Known proof gap: no live browser session against a multi-agent gateway; covered here by focused UI/controller tests plus gateway aggregation tests.

@steipete
steipete merged commit 99bd275 into openclaw:main May 29, 2026
101 checks passed
@Haderach-Ram Haderach-Ram mentioned this pull request May 29, 2026
Open
@syfvb syfvb mentioned this pull request Jun 3, 2026
Closed
@TurboTheTurtle TurboTheTurtle mentioned this pull request Jun 3, 2026
Merged
@LiuwqGit LiuwqGit mentioned this pull request Jun 5, 2026
Closed
This was referenced Jun 12, 2026
Closed
Merged
@clawsweeper clawsweeper Bot mentioned this pull request Jun 19, 2026
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

@steipete steipete

Labels

app: web-ui App: web-ui docs Improvements or additions to documentation gateway Gateway runtime merge-risk: 🚨 compatibility 🚨 May break existing users, config, migrations, defaults, or upgrade paths. merge-risk: 🚨 security-boundary 🚨 May affect sandboxing, authorization, credentials, or sensitive data. P2 Normal backlog priority with limited blast radius. proof: override Maintainer override for the external PR real behavior proof gate. rating: 🦐 gold shrimp Decent PR readiness signal, but merge confidence is limited. scripts Repository scripts size: L

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

[Bug]: Usage page agent filter broken since v2026.5.17 – sessions.usage API always scoped to "main" agent

3 participants

@Alix-007 @BingqingLyu @steipete