Site Behavior Lab

Baseline visit v2 schema (r2)

roblox.com

https://www.roblox.com/

Aug 14, 2026, 7:05 PM UTC

cut short: fingerprinting evidence was censored before completion — the in-page fingerprint observer did not finish (recorded loss count: 1); detector output evidence was censored before completion — 1 privacy-policy visit did not produce usable evidence; recorded by the scanner

Public corrections ledger · SBL-CORR-2026-001

This report has a reviewed clarification

SBL-CORR-2026-001: The historical input-probe disclosure overstates coverage. Unload-only transmissions were not reliably recorded. Focus and blur handlers could run, and per-field failures were not fully counted. These reports cannot establish absence of input transmission outside their retained observations.

Read the public review record

Evidence quality

This report is built on an incomplete visit

Incomplete evidence: evidence was cut short before completion (fingerprinting evidence was censored before completion — the in-page fingerprint observer did not finish (recorded loss count: 1); detector output evidence was censored before completion — 1 privacy-policy visit did not produce usable evidence). Counts for the affected evidence are lower bounds; families that completed carry only their ordinary limits. The evidence receipt states the exact per-visit quality.

What this actually means

roblox.com contacted 1 distinct catalogued tracking-related service during this visit.

SBL-CORR-2026-001: The historical input-probe disclosure overstates coverage. Unload-only transmissions were not reliably recorded. Focus and blur handlers could run, and per-field failures were not fully counted. These reports cannot establish absence of input transmission outside their retained observations.

1tracking-service entity
164cross-site requests

Plain-Language Findings

What this visit means

Unfamiliar terms are defined in the glossary
playwright-chromium

Context

Bottom line: this visit has review-worthy signals

The scan observed signals a non-expert should not have to decode from raw request tables. Some detector evidence was incomplete; each affected detector card states its own retained or unavailable scope, while completed request, cookie, and storage measurements keep their recorded exactness.

The cards below translate the evidence into plain language. Where a measured distribution exists, severity ranks this visit against percentiles from the report's exact schema, methodology, tracker-catalog, ServiceRole-taxonomy, metric-contract, producer, and Global Privacy Control cohort, with each percentile card naming its metric-specific measured-site denominator (among 98 sites with a successful load across all cohorts; other methodologies, request-capped visits, and post-choice consent visits are excluded from this denominator, while failed or block-page attempts are outside that coverage), a curated set of popular, mostly commercial sites, not a random sample of the web, and otherwise uses fixed reference thresholds. The request log, domain table, and methodology remain below for verification.

199 requests observed in one controlled visit.

Context

5 of 184 requests the engine evaluated matched Brave Shields filter lists

5 requests matched the default filter lists of Brave Shields, the ad and tracker blocker built into the Brave browser, while loading normally.

Computed with Brave's own ad-block engine and default filter lists in classification mode: matched requests were not blocked by the scanner and remain in this report's observed request counts. Matching shows what Shields would target on this visit's traffic; an actual Shields visit blocks these and also prevents their follow-on requests, so this number is neither a measured block count nor the total effect.

The hand-curated service catalog separately matched 11 request rows in this visit.Show matched requests

Context

This GPC comparison supports no comparable delta

The pair did not record the network egress region for both visits, and an unrecorded condition never counts as matching.

The raw request logs of both visits remain below for transparency, but the diff between them supports no claim about what the compared condition changed.

1 disqualifying condition named by the report's claim policy.

Context

Requests were dispatched to catalogued service domains

Cloudflare appeared in the request log.

A catalog match identifies a maintainer-reviewed service/domain mapping; it does not establish why an individual request occurred, what it carried, or whether profiling happened. Functional catalog labels include web analytics. This scan could not identify 2 of the 5 third-party hosts recorded here, so it cannot say who operates them. That is a limit of identity coverage, not evidence about the site. The role taxonomy assigns these additional services explicitly non-tracking operational, support, security, consent-management, or hosting roles: HUMAN Security and Sentry.

164 cross-site requests across 5 third-party hosts.Below the median for third-party domains across the 54 sites measured for this metric. Modest distinct tracking-service entities count.Show third-party requests

Nothing observed

No requests to major-platform domains were recorded

No requests to catalogued Google, Meta, TikTok, X, Microsoft, LinkedIn, or Pinterest domains were observed in this visit.

Major-platform domains were not observed in this single passive visit; interaction-gated requests could still load for real users.

5 cross-site domains seen overall.Show catalog-matched requests

Nothing observed

No Google Analytics observed

This visit did not contact Google Analytics.

Neither Google Analytics nor its remarketing sync was observed in this visit.

No google-analytics.com or googletagmanager.com requests.Show catalog-matched requests

Nothing observed

No third-party cookies observed

The automated visit did not observe third-party cookies.

This does not prove the site never uses cookies; it means this visit did not observe third-party cookies.

10 cookies total in this report.No third-party cookies observed.

Context

Incomplete evidence: No fingerprint-like API calls observed in retained evidence before collection stopped

The scan did not observe the instrumented high-entropy browser APIs.

This is an observation layer, not proof that fingerprinting is impossible. Evidence collection was cut short, so this covers only what was recorded before the cutoff.

Retained incomplete evidence includes 0 API event records; no exact API-family total is available.

Raw evidence

Open the interactive evidence explorer

The findings above are built from the stored report. Load the request rows, domains, cookies, storage keys, browser signals, visit phases, charts and CSV export when you want to check them; the versioned report JSON is always available below without this interface.

This versioned report is currently retained in the public corpus. Follow the currently retained site history, repeat the same public route when it is available, or request a transparent evidence correction.

Report ID
20260814-ca28b4ecdbc1457b46aac643fcc2e945
Site
roblox.com
Latest visit
Schema
v2 schema (r2)
Verify recorded provenance

Baseline visit

Started
Run quality
cut short: fingerprinting evidence was censored before completion — the in-page fingerprint observer did not finish (recorded loss count: 1); detector output evidence was censored before completion — 1 privacy-policy visit did not produce usable evidence; recorded by the scanner
Scanner
playwright-chromium
Browser
151.0.7922.34
Redaction
version 4
Methodology
shields-request-context-v2-adblock-rust-0.13.2-request-method-v1-playwright-1.62.1+subject-validity-v2+detector-coverage-v2+phase-kernel-v2+boundary-state-v1+consent-r2-v4+resource-budget-v1+proxy-traffic-v1+service-worker-block-v1+detector-accountability-v1+service-role-taxonomy-v1
Detector registry
node-detectors-v4100de91713270067dff4f5ecebeea61d330982c7a5aa33395bae3dd604adedd2
Tracker catalog digest
e94970de235fc80254de8ed99b94316a252e52aa1c2e748c8fbfc3c093b908f4
Brave-list instrument
engine adblock-rust-0.13.20f52cc524a8999a1022279aca74003ece9720f267bc678430727b7312d43f0ac
Normalization
redaction-v4+allowlists-v3:269f631f04090ce582644ee3cf0e5c5b6bb425dc4929bc283607b808bc9322a9+public-string-policy-v3:6c78c05523e1f16c88264d0144af33587bd6dc11e04d337a6af2d58190639266+tldts@7.4.10+node-evidence-policy-v1+r2-http-status-compat-v1

Variant visit

Started
Run quality
cut short: fingerprinting evidence was censored before completion — the in-page fingerprint observer did not finish (recorded loss count: 1); detector output evidence was censored before completion — 1 privacy-policy visit did not produce usable evidence; recorded by the scanner
Scanner
playwright-chromium
Browser
151.0.7922.34
Redaction
version 4
Methodology
shields-request-context-v2-adblock-rust-0.13.2-request-method-v1-playwright-1.62.1+subject-validity-v2+detector-coverage-v2+phase-kernel-v2+boundary-state-v1+consent-r2-v4+resource-budget-v1+proxy-traffic-v1+service-worker-block-v1+detector-accountability-v1+service-role-taxonomy-v1
Detector registry
node-detectors-v4100de91713270067dff4f5ecebeea61d330982c7a5aa33395bae3dd604adedd2
Tracker catalog digest
e94970de235fc80254de8ed99b94316a252e52aa1c2e748c8fbfc3c093b908f4
Brave-list instrument
engine adblock-rust-0.13.20f52cc524a8999a1022279aca74003ece9720f267bc678430727b7312d43f0ac
Normalization
redaction-v4+allowlists-v3:269f631f04090ce582644ee3cf0e5c5b6bb425dc4929bc283607b808bc9322a9+public-string-policy-v3:6c78c05523e1f16c88264d0144af33587bd6dc11e04d337a6af2d58190639266+tldts@7.4.10+node-evidence-policy-v1+r2-http-status-compat-v1

Recorded provenance is self-reported measurement metadata, not a cryptographic attestation. The public versioned report JSON contains the retained public evidence; see the methodology for interpretation and retention limits.

This report covers what the scanner measures, not everything a site can do. 16 browser surfaces are not instrumented at all (Every page except the one that was scanned; Anything that only loads after scrolling or interaction; Why a request happened, as a causal chain; and 13 others), 3 capabilities are declined by policy, and 3 are outside what any single visit can see. Their absence from this report is not evidence that they did not occur. Read the published coverage boundary.

Detector accuracy is unmeasured. No calibration study currently publishes a precision or recall for the detectors on this page, so a detector match here is a recorded observation, not a calibrated classification. How detector accuracy is measured.

Approved use boundary: This report is investigative evidence that requires independent corroboration. It is not a standalone legal determination and not the sole exhibit in a legal proceeding. Recorded in this project's release manifest on .