fix(voicecall): redact read-scoped status payloads by eleqtrizit · Pull Request #97870 · openclaw/openclaw

eleqtrizit · 2026-06-29T16:19:56Z

Summary

Redacts read-scoped voice-call status payloads so status callers receive a purpose-built call summary instead of raw live call records.

Changes

Adds a local voice-call status DTO projection for voicecall.status responses.
Applies the projection to single-call status, active-call list status, and the voice-call tool status path.
Omits phone numbers, session routing keys, transcripts, processed event ids, and raw metadata from status payloads while preserving status identifiers and lifecycle fields.
Adds regression coverage for single-call, list, and tool status redaction.

Validation

corepack pnpm format:fix extensions/voice-call/index.ts extensions/voice-call/index.test.ts
corepack pnpm test extensions/voice-call/index.test.ts
.agents/skills/autoreview/scripts/autoreview --mode local

eleqtrizit · 2026-06-29T16:22:08Z

Real behavioral proof on the current PR head (ab0094cf090d95a6760e496fd694180ec0bc95c8):

Runtime setup

Started the real OpenClaw Gateway from this head with an isolated state directory and the bundled voice-call plugin configured with its network-free mock provider.
Created an actual active call through the running Gateway's write-scoped voicecall.initiate RPC, supplying dummy phone endpoints, a session routing value, an initial message, and requester metadata.
Confirmed through openclaw voicecall tail --since 1 that the persisted internal call record contains from, to, sessionKey, transcript, processedEventIds, and metadata. This proves the test call exercised a real sensitive CallRecord; the source data was not pre-redacted or replaced by a fixture.

Read-scope boundary proof

A separate WebSocket client connected to that running Gateway with exactly scopes: ["operator.read"]:

{
  "requestedScopes": ["operator.read"],
  "list": {
    "found": true,
    "calls": [{
      "callId": "fff63af6-a36e-4f86-9120-65df52a84955",
      "providerCallId": "mock-fff63af6-a36e-4f86-9120-65df52a84955",
      "provider": "mock",
      "direction": "outbound",
      "state": "initiated",
      "startedAt": 1782753710598
    }]
  },
  "single": {
    "found": true,
    "call": {
      "callId": "fff63af6-a36e-4f86-9120-65df52a84955",
      "providerCallId": "mock-fff63af6-a36e-4f86-9120-65df52a84955",
      "provider": "mock",
      "direction": "outbound",
      "state": "initiated",
      "startedAt": 1782753710598
    }
  },
  "writeControl": "missing scope: operator.write"
}

That same read-only client was denied voicecall.end, proving the session was genuinely read-scoped rather than an admin/shared-secret session whose broader permissions masked the boundary.

Observed behavior:

The list envelope remains { found: true, calls: [...] }.
The single-call envelope remains { found: true, call: ... }.
Operational identifiers and lifecycle fields remain available.
Phone endpoints, session routing data, transcript/event arrays, and raw metadata remain in the canonical internal record but do not cross the read-scoped status boundary.

This also separates the compatibility concern from the completed-call/history use case: detailed records still exist in voice-call storage, so a future privileged or history-specific detail surface can expose them intentionally without widening operator.read status again.

Validation

node scripts/run-vitest.mjs extensions/voice-call/index.test.ts — 1 file passed, 39 tests passed.
git diff --check origin/main...HEAD — passed.
Exact-head CI is green for extension tests, lint, production/test types, plugin boundaries, and security checks.
Exact-head Real behavior proof check: https://github.com/openclaw/openclaw/actions/runs/28389481257/job/84112567907

No carrier or external telephony service was used; the proof exercised the real Gateway, plugin loader, plugin runtime, RPC authorization, call manager, and persistent call store with the bundled mock provider.

clawsweeper · 2026-06-29T16:22:13Z

Codex review: needs maintainer review before merge. Reviewed June 29, 2026, 1:28 PM ET / 17:28 UTC.

Summary
The PR adds a local voice-call status projection and uses it for gateway and tool status responses so read-scoped callers no longer receive raw CallRecord fields.

PR surface: Source +33, Tests +80. Total +113 across 2 files.

Reproducibility: yes. at source level: current main and v2026.6.10 return raw CallRecord objects from read-scoped voice-call status paths. I did not place a live call in this read-only review, but the PR discussion includes exact-head live Gateway proof.

Review metrics: 1 noteworthy metric.

Status response paths: 4 projected/redacted. Gateway single-call status, gateway active-call list status, action get_status, and legacy mode: status now share one narrowed payload contract that maintainers should notice before merge.

Stored data model
Persistent data-model change detected: vector/embedding metadata: extensions/voice-call/index.test.ts. Confirm migration or upgrade compatibility proof before merge.

Merge readiness
Overall: 🐚 platinum hermit
Proof: 🦞 diamond lobster
Patch quality: 🐚 platinum hermit
Result: ready for maintainer review.

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

Rank-up moves:

[P2] Maintainers should explicitly accept the redacted operator.read status contract or request a separate privileged raw-detail/history surface before merge.

Risk before merge

[P1] Merging intentionally removes shipped raw CallRecord fields from voicecall.status and voice_call status responses, so existing read/operator clients that parse phone numbers, session keys, transcripts, processed event ids, or metadata from those status payloads will stop receiving them.
[P1] The open persisted-status request at [voice-call] get_status returns found:false for completed calls once evicted from in-memory manager — no fallback to persisted call store #96586 still expects completed-call status to retrieve transcript-bearing persisted records, so maintainers should decide whether that future need belongs in a separate privileged or history-specific surface.

Maintainer options:

Accept The Redacted Status Contract
Merge as-is only if maintainers decide operator.read voice-call status should return a limited call summary instead of raw CallRecord objects.
Add A Privileged Raw-Detail Surface
If operators or persisted-history work still need phone numbers, transcripts, routing keys, or metadata, add a separate admin/write-scoped or history-specific detail surface before removing those fields from read-scoped status.
Pause For API Migration Direction
Hold the PR if maintainers want caller audit, documentation, or a migration/deprecation path before narrowing the shipped status payload.

Next step before merge

[P2] The remaining blocker is maintainer acceptance of a shipped response-shape narrowing under the protected maintainer label, not a narrow automated code repair.

Security
Cleared: The diff reduces sensitive voice-call status exposure and does not add dependencies, scripts, permissions, secret handling, lockfile changes, or new supply-chain surface.

Review details

Best possible solution:

Land the DTO projection only after maintainers explicitly accept the narrowed read-scoped status contract; if raw records remain needed, expose them through a separate privileged/admin or history-specific surface.

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

Yes, at source level: current main and v2026.6.10 return raw CallRecord objects from read-scoped voice-call status paths. I did not place a live call in this read-only review, but the PR discussion includes exact-head live Gateway proof.

Is this the best way to solve the issue?

Yes technically: projecting a local DTO at the status egress points is the narrowest maintainable fix for the read-scope exposure. The unresolved part is maintainer acceptance of the shipped compatibility break and whether raw details need a separate privileged/history surface.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against 6de357ad4772.

Label changes

Label changes:

add proof: sufficient: Contributor real behavior proof is sufficient. The PR discussion includes exact-head live Gateway proof with a mock provider, an operator.read client receiving redacted status payloads, a denied write call, and a successful exact-head Real behavior proof job.

Label justifications:

P2: This is focused voice-call privacy hardening with limited file scope, but it changes a shipped status API shape.
merge-risk: 🚨 compatibility: The PR removes raw CallRecord fields from shipped status responses, which can break callers relying on the previous payload shape.
rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🦞 diamond lobster and patch quality is 🐚 platinum hermit.
status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (live_output): The PR discussion includes exact-head live Gateway proof with a mock provider, an operator.read client receiving redacted status payloads, a denied write call, and a successful exact-head Real behavior proof job.
proof: sufficient: Contributor real behavior proof is sufficient. The PR discussion includes exact-head live Gateway proof with a mock provider, an operator.read client receiving redacted status payloads, a denied write call, and a successful exact-head Real behavior proof job.

Evidence reviewed

PR surface:

Source +33, Tests +80. Total +113 across 2 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	1	37	4	+33
Tests	1	82	2	+80
Docs	0	0	0	0
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	2	119	6	+113

What I checked:

Repository policy applied: Root and scoped extension policies were read; the review applied the plugin-boundary and compatibility guidance for shipped response-shape changes. (AGENTS.md:1, 6de357ad4772)
Current main returns raw call records: On current main, voicecall.status returns rt.manager.getActiveCalls() for list responses and returns the raw call object for single-call responses. (extensions/voice-call/index.ts:618, 6de357ad4772)
Current tool status returns raw call records: On current main, the voice_call get_status and legacy mode: "status" tool paths return the raw call object. (extensions/voice-call/index.ts:761, 6de357ad4772)
Raw record contains sensitive fields: CallRecordSchema includes from, to, sessionKey, transcript, processedEventIds, and metadata, matching the fields this PR removes from read-scoped status payloads. (extensions/voice-call/src/types.ts:152, 6de357ad4772)
Latest release has the same raw status behavior: The latest release tag still returns raw active-call and single-call/tool status objects, so this PR changes shipped behavior rather than only unreleased main behavior. (extensions/voice-call/index.ts:618, aa69b12d0086)
PR head projects the status DTO at all status egress points: The PR adds toVoiceCallStatus and applies it to gateway list status, gateway single-call status, get_status, and legacy mode: "status". (extensions/voice-call/index.ts:236, ab0094cf090d)

Likely related people:

steipete: GitHub path history shows the original voice-call parity restoration and repeated voice-call plugin, status, CLI, state, and test maintenance. (role: feature-history owner; confidence: high; commits: 42c17adb5e4d, 4d4748e8071a, 39a931f1bfe0; files: extensions/voice-call/index.ts, extensions/voice-call/index.test.ts, extensions/voice-call/src/types.ts)
scoootscooob: Recent history shows scoped voice-call gateway method work and realtime voice-call changes near the same plugin status/control boundary. (role: adjacent voice-call contributor; confidence: medium; commits: 0c1df3531599, 79dd65e20889, b2f21853489d; files: extensions/voice-call/index.ts, extensions/voice-call/index.test.ts)
mushuiyu886: Recent merged voice-call work changed canonical session-key behavior adjacent to the session routing fields being removed from read-scoped status payloads. (role: recent area contributor; confidence: medium; commits: 65fec9d787e3; files: extensions/voice-call/src/manager.ts, extensions/voice-call/src/webhook.ts, extensions/voice-call/src/config.ts)
Alix-007: Local blame points the current status and schema lines to the grafted checkout commit, but this is weaker than GitHub path history because the checkout history is collapsed. (role: current-checkout blame signal; confidence: low; commits: 5f86c3a90f69; files: extensions/voice-call/index.ts, extensions/voice-call/src/types.ts)

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 · 2026-06-29T16:22:15Z

🦞🧹
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.

clawsweeper · 2026-06-29T17:03:20Z

🦞🧹
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.

eleqtrizit · 2026-06-29T17:23:14Z

@clawsweeper re-review

* fix(voicecall): redact read-scoped status payloads * fix(voicecall): make status assertion lint-safe

* fix(voicecall): redact read-scoped status payloads * fix(voicecall): make status assertion lint-safe (cherry picked from commit 825aafa)

fix(voicecall): redact read-scoped status payloads

06d8b8b

openclaw-barnacle Bot added channel: voice-call Channel integration: voice-call size: S maintainer Maintainer-authored PR labels Jun 29, 2026

fix(voicecall): make status assertion lint-safe

ab0094c

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 29, 2026

eleqtrizit merged commit 825aafa into openclaw:main Jun 29, 2026
120 of 126 checks passed

This was referenced Jun 30, 2026

fix #96586: [Bug]: voice-call get_status returns found:false for completed calls evicted from in-memory manager #96727

Closed

fix(voice-call): fall back to persisted call store on status misses #96624

Closed

github-actions Bot pushed a commit to Desicool/openclaw that referenced this pull request Jun 30, 2026

fix(voicecall): redact read-scoped status payloads (openclaw#97870)

0079aae

* fix(voicecall): redact read-scoped status payloads * fix(voicecall): make status assertion lint-safe

eleqtrizit deleted the 784 branch June 30, 2026 19:27

QiuYuang pushed a commit to QiuYuang/openclaw that referenced this pull request Jul 1, 2026

fix(voicecall): redact read-scoped status payloads (openclaw#97870)

e493b25

* fix(voicecall): redact read-scoped status payloads * fix(voicecall): make status assertion lint-safe

chenyangjun-xy pushed a commit to chenyangjun-xy/openclaw that referenced this pull request Jul 1, 2026

fix(voicecall): redact read-scoped status payloads (openclaw#97870)

ae77c65

* fix(voicecall): redact read-scoped status payloads * fix(voicecall): make status assertion lint-safe

chenyangjun-xy pushed a commit to chenyangjun-xy/openclaw that referenced this pull request Jul 3, 2026

fix(voicecall): redact read-scoped status payloads (openclaw#97870)

f25ebee

* fix(voicecall): redact read-scoped status payloads * fix(voicecall): make status assertion lint-safe

NiTeCoMM-code mentioned this pull request Jul 3, 2026

[voice-call] get_status returns found:false for completed calls once evicted from in-memory manager — no fallback to persisted call store #96586

Closed

clawsweeper Bot mentioned this pull request Jul 4, 2026

fix(voice-call): resolve completed calls from the persisted store on status misses #99797

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(voicecall): redact read-scoped status payloads#97870

fix(voicecall): redact read-scoped status payloads#97870
eleqtrizit merged 2 commits into
openclaw:mainfrom
eleqtrizit:784

eleqtrizit commented Jun 29, 2026 •

edited

Loading

Uh oh!

eleqtrizit commented Jun 29, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Jun 29, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Jun 29, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Jun 29, 2026 •

edited

Loading

Uh oh!

eleqtrizit commented Jun 29, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Uh oh!

Conversation

eleqtrizit commented Jun 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes

Validation

Uh oh!

eleqtrizit commented Jun 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Runtime setup

Read-scope boundary proof

Validation

Uh oh!

clawsweeper Bot commented Jun 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clawsweeper Bot commented Jun 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clawsweeper Bot commented Jun 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

eleqtrizit commented Jun 29, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

eleqtrizit commented Jun 29, 2026 •

edited

Loading

eleqtrizit commented Jun 29, 2026 •

edited

Loading

clawsweeper Bot commented Jun 29, 2026 •

edited

Loading

clawsweeper Bot commented Jun 29, 2026 •

edited

Loading

clawsweeper Bot commented Jun 29, 2026 •

edited

Loading