Disable Chrome MCP telemetry watchdog by default by rohitjavvadi · Pull Request #85886 · openclaw/openclaw

rohitjavvadi · 2026-05-24T01:04:54Z

Summary

Pass --no-usage-statistics when OpenClaw launches chrome-devtools-mcp@latest.
Preserve explicit profile-level mcpArgs overrides for usage statistics.
Add regression coverage for the default arg list, explicit opt-in, explicit opt-out dedupe, and custom-command launches.

Motivation / root cause

chrome-devtools-mcp enables usage statistics by default and can spawn a detached watchdog process. OpenClaw closes the MCP SDK stdio transport, but that only owns the direct child process. Disabling usage statistics at launch prevents the detached telemetry watchdog from being created in the default OpenClaw browser automation path.

Refs #85721

Real behavior proof

Behavior addressed: Browser automation launches Chrome DevTools MCP without leaving a detached usage-statistics watchdog behind after the MCP client is closed.

Real environment tested: macOS local checkout, Node 24.10.0, real npx chrome-devtools-mcp@latest process launched through @modelcontextprotocol/sdk StdioClientTransport.

Exact steps or command run after this patch: Started a real MCP SDK Client with StdioClientTransport({ command: "npx", args: buildChromeMcpArgs({ cdpUrl: "http://127.0.0.1:9" }) }), called client.connect(), client.listTools(), then client.close(), waited 1.2s, and scanned ps axo pid=,ppid=,command= for chrome-devtools-mcp and watchdog/main.js.

Evidence after fix: Copied live console output from the real SDK stdio close/process-scan smoke:

{
  "generatedArgs": [
    "-y",
    "chrome-devtools-mcp@latest",
    "--browserUrl",
    "http://127.0.0.1:9",
    "--no-usage-statistics",
    "--experimentalStructuredContent",
    "--experimental-page-id-routing"
  ],
  "transportPid": 14467,
  "remainingRelevantProcesses": [],
  "watchdogProcessesRemaining": 0
}

Observed result after fix: Live output shows --no-usage-statistics in the generated argv, remainingRelevantProcesses: [], and watchdogProcessesRemaining: 0; no chrome-devtools-mcp or telemetry watchdog process remained after closing the SDK client.

What was not tested: I did not run a full Chrome browser UI workflow or a Linux remote/Testbox smoke in this pass.

Regression tests

node scripts/run-vitest.mjs extensions/browser/src/browser/chrome-mcp.test.ts
node scripts/run-vitest.mjs extensions/browser/src/browser/server-context.existing-session.test.ts extensions/browser/src/browser/server-context.stop-running-browser.test.ts extensions/browser/src/browser/routes/basic.existing-session.test.ts

Exact commands run

./node_modules/.bin/oxfmt --check extensions/browser/src/browser/chrome-mcp.ts extensions/browser/src/browser/chrome-mcp.test.ts
git diff --check origin/main...HEAD
node scripts/run-vitest.mjs extensions/browser/src/browser/chrome-mcp.test.ts
node scripts/run-vitest.mjs extensions/browser/src/browser/server-context.existing-session.test.ts extensions/browser/src/browser/server-context.stop-running-browser.test.ts extensions/browser/src/browser/routes/basic.existing-session.test.ts
Real MCP SDK stdio close/process-scan smoke described above.

User-visible behavior

Existing-session browser automation should no longer accumulate detached Chrome DevTools MCP telemetry watchdog processes during normal OpenClaw-managed MCP launches.

Compatibility

Users who explicitly want upstream usage statistics can still pass a usage-statistics arg in browser.profiles.*.mcpArgs; explicit args override OpenClaw's default opt-out.

Risks

Low. The change adds one documented Chrome DevTools MCP flag to the existing argv list and keeps profile-level custom args as separate argv entries.

clawsweeper · 2026-05-24T01:07:11Z

Codex review: passed.

Latest ClawSweeper review: 2026-05-24 01:59 UTC / May 23, 2026, 9:59 PM ET.

Workflow note: Future ClawSweeper reviews update this same comment in place.

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.

PR Surface
Source +7, Tests +36, Docs +1. Total +44 across 3 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	1	8	1	+7
Tests	1	37	1	+36
Docs	1	1	0	+1
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	3	46	2	+44

Summary
The PR adds the Chrome DevTools MCP --no-usage-statistics default launch arg, honors explicit profile usage-statistics mcpArgs, adds regression tests, and adds a changelog entry.

Reproducibility: yes. source-reproducible: current main builds Chrome MCP launch args without the upstream opt-out flag, while upstream defaults usage statistics on and starts a detached watchdog when telemetry is initialized. I did not run a fresh failing current-main process leak loop in this read-only review.

PR rating
Overall: 🦞 diamond lobster
Proof: 🦞 diamond lobster
Patch quality: 🦞 diamond lobster
Summary: Focused bug fix with strong dependency-contract evidence, targeted tests, and real process-level proof; no blocking findings found.

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.

Real behavior proof
Sufficient (live_output): The PR body provides sufficient after-fix live output from a real MCP SDK stdio close/process-scan smoke showing no remaining MCP or watchdog processes.

Risk before merge

The supplied real behavior proof is a macOS SDK stdio process-scan smoke rather than a full Linux/Chrome UI loop, so exact-head CI and any available browser lane remain useful before final merge.

Maintainer options:

Decide the mitigation before merge
Land this narrow arg-builder fix after exact-head checks, keeping explicit profile args as the operator opt-in path for upstream usage statistics.
Pause or close
Do not merge this PR until maintainers decide whether the risk is worth taking.

Next step before merge
No repair lane is needed; the automerge-armed branch has no actionable review findings, so exact-head checks and mergeability should gate it.

Security
Cleared: No security or supply-chain regression was found; the diff only changes Chrome MCP argv construction, focused tests, and the changelog.

Review details

Best possible solution:

Land this narrow arg-builder fix after exact-head checks, keeping explicit profile args as the operator opt-in path for upstream usage statistics.

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

Yes, source-reproducible: current main builds Chrome MCP launch args without the upstream opt-out flag, while upstream defaults usage statistics on and starts a detached watchdog when telemetry is initialized. I did not run a fresh failing current-main process leak loop in this read-only review.

Is this the best way to solve the issue?

Yes; adding the documented upstream opt-out to OpenClaw's default Chrome MCP argv while removing it only for explicit profile usage-statistics args is the narrow maintainable fix.

Label justifications:

P1: The linked browser automation leak can accumulate detached Chrome DevTools MCP watchdog processes and disrupt real browser-agent workflows through memory pressure.
rating: 🦞 diamond lobster: Current PR rating is 🦞 diamond lobster because proof is 🦞 diamond lobster, patch quality is 🦞 diamond lobster, and Focused bug fix with strong dependency-contract evidence, targeted tests, and real process-level proof; no blocking findings found.
status: 🚀 automerge armed: This PR is in ClawSweeper's automerge lane. Sufficient (live_output): The PR body provides sufficient after-fix live output from a real MCP SDK stdio close/process-scan smoke showing no remaining MCP or watchdog processes.
proof: sufficient: Contributor real behavior proof is sufficient. The PR body provides sufficient after-fix live output from a real MCP SDK stdio close/process-scan smoke showing no remaining MCP or watchdog processes.

What I checked:

Current main lacks the opt-out: At current main, DEFAULT_CHROME_MCP_FEATURE_ARGS contains only the structured-content and page-id-routing flags, and buildChromeMcpArgsFromOptions appends those defaults before profile extraArgs; no usage-statistics opt-out is emitted by default. (extensions/browser/src/browser/chrome-mcp.ts:74, 12f82270cf07)
PR head implements the default opt-out: PR head adds --no-usage-statistics to the default feature args, removes that default when profile args explicitly mention usage-statistics, and then appends the explicit profile args. (extensions/browser/src/browser/chrome-mcp.ts:75, 68249b1f5866)
Regression coverage: PR head tests cover default auto-connect, browserUrl, wsEndpoint, custom command launches, explicit --usage-statistics opt-in, and explicit --no-usage-statistics dedupe. (extensions/browser/src/browser/chrome-mcp.test.ts:162, 68249b1f5866)
Dependency contract: Upstream Chrome DevTools MCP README and CLI options state usage statistics are enabled by default, can be disabled with --no-usage-statistics, and are also disabled by CHROME_DEVTOOLS_MCP_NO_USAGE_STATISTICS or CI.
Telemetry watchdog behavior: Upstream initializes ClearcutLogger only when serverArgs.usageStatistics is true, and WatchdogClient starts watchdog/main.js as a detached, unref'd child process.
Current main and release status: The PR head is not an ancestor of current main and no local branch or tag contains it, so this fix is still a PR-only change rather than implemented on current main or the latest release. (68249b1f5866)

Likely related people:

steipete: Peter Steinberger introduced the Chrome MCP existing-session docs, moved the implementation into the bundled browser plugin, and owns the blamed current arg-builder lines sampled for this review. (role: feature-history owner; confidence: high; commits: 593964560ba7, 8eeb7f082975, d73f3ac85d5f; files: extensions/browser/src/browser/chrome-mcp.ts, extensions/browser/src/browser/chrome-mcp.test.ts, docs/tools/browser.md)
vincentkoc: Recent browser-plugin history includes adjacent cycle/runtime seam work in the same browser area, making Vincent a useful secondary routing candidate if broader browser plugin context is needed. (role: adjacent recent contributor; confidence: medium; commits: 0e54440ecc39; files: extensions/browser/src/browser/chrome-mcp.ts)

Codex review notes: model gpt-5.5, reasoning high; reviewed against 12f82270cf07.

clawsweeper · 2026-05-24T01:17:44Z

ClawSweeper PR egg

✨ Hatched: 🥚 common Sunspot Diff Drake

Hatch command

Comment @clawsweeper hatch when this PR is hatchable.

Hatchability rules:

Merged PRs are hatchable.
Open PRs are hatchable when they are status: 👀 ready for maintainer look, status: 🚀 automerge armed, or labeled clawsweeper:automerge.
Closed unmerged PRs are hatchable only when one of those hatchable labels is still present in the durable record.

Rarity: 🥚 common.
Trait: sleeps inside passing CI.
Image traits: location branch lighthouse; accessory rollback rope; palette rose quartz and slate; mood focused; pose guarding a tiny green check; shell woven fiber shell; lighting golden review-room light; background small review tokens.
Share on X: post this hatch
Copy: My PR egg hatched a 🥚 common Sunspot Diff Drake in ClawSweeper.

What is this egg doing here?

Eggs appear after the PR passes real-behavior proof. It is here for vibes, not verdicts: it does not change labels, ratings, merge decisions, or automation.
The shell reacts to review momentum: open follow-up work warms it up, re-review makes it wobble, and a clean final review lets it hatch.
Hatchability usually comes from sufficient real-behavior proof, no blocking P0/P1/P2 findings, no security attention needed, and clean correctness. A merged PR is already final, so merge makes the egg hatchable independently.
The hatch is seeded from this repository and PR number, so the same PR keeps the same creature; the reviewed head SHA can only change safe visual details.
Rarity is just collectible sparkle: 🥚 common, 🌱 uncommon, 💎 rare, ✨ glimmer, and 🌈 legendary.

Takhoffman · 2026-05-24T01:28:15Z

@clawsweeper automerge

clawsweeper · 2026-05-24T01:28:18Z

🦞🧹
ClawSweeper automerge is enabled.

Head: 68249b1f5866
Label: clawsweeper:automerge
Action: exact-head review queued (workflow sweep.yml, event repository_dispatch).
Flow: review this head, repair/rebase only if needed, then re-review the exact repaired head before merge.

Draft PRs stay fix-only until GitHub marks them ready for review. Pause with /clawsweeper stop.

Automerge progress:

2026-05-24 01:28:15 UTC review queued fd697db81be8 (queued)

2026-05-24 01:29:41 UTC repair queued fd697db81be8 (autonomous) Run: https://github.com/openclaw/clawsweeper/actions/runs/26348578714

2026-05-24 01:31:49 UTC repair started (running) in 0s Run: https://github.com/openclaw/clawsweeper/actions/runs/26348578714 automerge-openclaw-openclaw-85886

2026-05-24 01:32:04 UTC validation plan (passed) in 15s Run: https://github.com/openclaw/clawsweeper/actions/runs/26348578714 pnpm check:changed; pnpm lint; pnpm check:test-types

2026-05-24 01:32:16 UTC Codex write preflight (passed) in 27s Run: https://github.com/openclaw/clawsweeper/actions/runs/26348578714 danger-full-access

2026-05-24 01:36:59 UTC Codex edit 1 5530f2dd5643 (complete) in 5m 10s Run: https://github.com/openclaw/clawsweeper/actions/runs/26348578714 exit 0

2026-05-24 01:40:36 UTC validation and review 1 68249b1f5866 (base moved) in 8m 47s Run: https://github.com/openclaw/clawsweeper/actions/runs/26348578714 rebased

2026-05-24 01:40:41 UTC repair completed 68249b1f5866 (branch updated) in 8m 52s Run: https://github.com/openclaw/clawsweeper/actions/runs/26348578714 initial automerge rebase is delegated to Codex repair

2026-05-24 01:40:40 UTC review queued 68249b1f5866 (after repair)

2026-05-24 01:50:43 UTC automerge wait 68249b1f5866 (waiting) in 18m 54s Run: https://github.com/openclaw/clawsweeper/actions/runs/26348578714 waiting for GitHub checks: auto-response:QUEUED, Real behavior proof:IN_PROGRESS

2026-05-24 02:02:27 UTC review passed 68249b1f5866 (structured ClawSweeper verdict: pass (sha=68249b1f586660471ac8c84d8c3f1834fd8c4...)

2026-05-24 01:50:44 UTC repair finished 68249b1f5866 (pushed) in 18m 55s Run: https://github.com/openclaw/clawsweeper/actions/runs/26348578714 repair_contributor_branch

2026-05-24 02:02:40 UTC merged 68249b1f5866 (merged by ClawSweeper automerge)

2026-05-24 02:02:44 UTC review queued 68249b1f5866 (queued)

rohitjavvadi · 2026-05-24T01:52:23Z

@clawsweeper re-review

clawsweeper · 2026-05-24T01:52:25Z

🦞🧹
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 targeted re-review finished, the durable review comment was updated, and the synced verdict was routed.
Run: https://github.com/openclaw/clawsweeper/actions/runs/26349014706
Updated: 2026-05-24T02:02:46.284Z

Summary: - The PR adds the Chrome DevTools MCP `--no-usage-statistics` default launch arg, honors explicit profile usage-statistics `mcpArgs`, adds regression tests, and adds a changelog entry. - Reproducibility: yes. source-reproducible: current main builds Chrome MCP launch args without the upstream o ... etry is initialized. I did not run a fresh failing current-main process leak loop in this read-only review. Automerge notes: - PR branch already contained follow-up commit before automerge: Disable Chrome MCP telemetry watchdog by default Validation: - ClawSweeper review passed for head 68249b1. - Required merge gates passed before the squash merge. Prepared head SHA: 68249b1 Review: openclaw#85886 (comment) Co-authored-by: Rohit <rohitjavvadi2@gmail.com> Co-authored-by: clawsweeper <274271284+clawsweeper[bot]@users.noreply.github.com> Co-authored-by: clawsweeper[bot] <274271284+clawsweeper[bot]@users.noreply.github.com> Approved-by: takhoffman Co-authored-by: takhoffman <781889+takhoffman@users.noreply.github.com>

openclaw-barnacle Bot added size: XS triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. labels May 24, 2026

openclaw-barnacle Bot added proof: supplied External PR includes structured after-fix real behavior proof. and removed triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. labels May 24, 2026

rohitjavvadi marked this pull request as ready for review May 24, 2026 01:08

clawsweeper Bot added the clawsweeper:automerge Maintainer opted this PR into bounded ClawSweeper-reviewed automerge label May 24, 2026

rohitjavvadi and others added 2 commits May 24, 2026 01:40

fix(browser): disable chrome mcp telemetry watchdog

206f9b3

Disable Chrome MCP telemetry watchdog by default

68249b1

clawsweeper Bot force-pushed the fix/chrome-mcp-disable-usage-watchdog branch from fd697db to 68249b1 Compare May 24, 2026 01:40

clawsweeper Bot added status: 🚀 automerge armed This PR is in ClawSweeper's automerge lane. and removed status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. labels May 24, 2026

clawsweeper Bot merged commit d581415 into openclaw:main May 24, 2026
101 of 103 checks passed

Haderach-Ram mentioned this pull request May 24, 2026

Ecosystem Digest — 2026-05-24 Haderach-Ram/openclaw-radar#17

Open

github-actions Bot mentioned this pull request May 24, 2026

📡 Upstream Digest — 2026-05-24 02:32 UTC curtismercier/openclaw-mods#930

Open

SebTardif mentioned this pull request May 28, 2026

fix: relay Claude CLI assistant partial messages as streaming deltas #86649

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Disable Chrome MCP telemetry watchdog by default#85886

Disable Chrome MCP telemetry watchdog by default#85886
clawsweeper[bot] merged 2 commits into
openclaw:mainfrom
rohitjavvadi:fix/chrome-mcp-disable-usage-watchdog

rohitjavvadi commented May 24, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented May 24, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented May 24, 2026

Uh oh!

Takhoffman commented May 24, 2026

Uh oh!

clawsweeper Bot commented May 24, 2026 •

edited

Loading

Uh oh!

rohitjavvadi commented May 24, 2026

Uh oh!

clawsweeper Bot commented May 24, 2026 •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

rohitjavvadi commented May 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Motivation / root cause

Real behavior proof

Regression tests

Exact commands run

User-visible behavior

Compatibility

Risks

Uh oh!

clawsweeper Bot commented May 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clawsweeper Bot commented May 24, 2026

Hatch command

Uh oh!

Takhoffman commented May 24, 2026

Uh oh!

clawsweeper Bot commented May 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rohitjavvadi commented May 24, 2026

Uh oh!

clawsweeper Bot commented May 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

rohitjavvadi commented May 24, 2026 •

edited

Loading

clawsweeper Bot commented May 24, 2026 •

edited

Loading

clawsweeper Bot commented May 24, 2026 •

edited

Loading

clawsweeper Bot commented May 24, 2026 •

edited

Loading