GCVE-1988-2026-0099
Vulnerability from gna-1988 – Published: 2026-09-07 13:20 – Updated: 2026-09-11 11:35| URL | Tags |
|---|---|
| https://vuln.freearchive.org/archive/full-disclos… | technical-descriptionexploit |
| https://seclists.org/fulldisclosure/2026/Jul/15 | technical-description |
| https://:443 | |
| https://nmap.org/mailman/listinfo/fulldisclosure | |
| https://seclists.org/fulldisclosure/ |
| Vendor | Product | Version | CPE status | |
|---|---|---|---|---|
| Laser Romae S.R.L. | Openblow |
Affected:
unknown
|
guessed |
{
"containers": {
"cna": {
"affected": [
{
"product": "Openblow",
"vendor": "Laser Romae\u202fS.R.L.",
"versions": [
{
"status": "affected",
"version": "unknown"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Red Nanaki via Fulldisclosure"
}
],
"descriptions": [
{
"lang": "en",
"value": "OpenBlow Multiple Deanonymization Vulnerabilities\n\nSummary\n\nA production deployment was observed (HTTP archive of a full, real whistleblower\nsubmission) to route its anonymous reporting flow through Google. The intake\nCAPTCHA is Google reCAPTCHA, enforced as a mandatory, server-validated gate\non report submission, and the UI additionally pulls a web font from\nfonts.gstatic.com. As a result, every prospective whistleblower\u0027s browser\nmakes direct connections to www.google.com and www.gstatic.com before a\nreport can be filed, disclosing the reporter\u0027s source IP, browser fingerprint,\nand the precise identity of the whistleblowing site to a third party. The\nabsence\nof a Content-Security-Policy and Referrer-Policy is what permits these\ncross-origin loads.\n\nThe defects, in severity order:\n\n1. (Critical) Google reCAPTCHA is a mandatory gate on the anonymous\nsubmission flow - a whistleblower cannot submit without first being exposed to\nGoogle.\n2. (High) A web font is loaded from fonts.gstatic.com, a second independent\nIP/fingerprint disclosure to Google.\n3. (Medium) No Content-Security-Policy and no Referrer-Policy on\nresponses, which is what allows the third-party loads above.\n\nComponent\n\ndeployment / configuration\n\nClassification\n\nvalid\n\nEvidence: HAR capture of one end-to-end flow (load intake \u2192 upload attachments \u2192\nPOST /api/v0/whistleblower/tip \u2192 retrieve). The submission could not complete\nwithout a Google reCAPTCHA token, which is present in the submission body and is\nproduced only after live calls to Google.\n\nDetails\n\nHosts contacted during a single anonymous submission:\n(first-party API + assets)\nwww.google.com (reCAPTCHA: api.js, api2/anchor, api2/reload [POST], \u2026)\nwww.gstatic.com (reCAPTCHA static bundle)\nfonts.gstatic.com (Roboto .woff2 web font)\n\nSubmission endpoint:\nPOST /api/v0/whistleblower/tip\nrequest body field \"captcha\" carries a Google reCAPTCHA token:\n\"captcha\":\"0cAFcWeA4BGSN1zd5E_vRIN\u2026\"\n-\u003e the token is required and server-validated: submission is gated on Google.\n\nSite identity leaked to Google (reCAPTCHA anchor \"co\" parameter, base64):\nco -\u003e decodes to https://:443\n\n1. (Critical) Mandatory Google reCAPTCHA on the anonymous intake path\n\nThe submission request carries a Google reCAPTCHA token in the captcha field of\nthe POST /api/v0/whistleblower/tip body, and that token can only be obtained\nafter the browser performs the reCAPTCHA handshake with Google (api.js \u2192\napi2/anchor \u2192 api2/reload). The flow therefore cannot proceed without\ncontacting Google.\n\nGoogle consequently receives, for every prospective reporter:\n\n- the source IP address of the whistleblower\u0027s browser;\n- a browser/device fingerprint (reCAPTCHA\u0027s purpose is behavioural/device\nscoring);\n- the exact whistleblowing site being used - the reCAPTCHA co parameter\nbase64-encodes the full origin (https://:443).\n\nThis contradicts the platform\u0027s anonymity guarantee, under which no third party\nmust be able to learn the identity of a source. It also degrades Tor usability:\nreCAPTCHA routinely blocks or challenges Tor exit nodes, pushing at-risk users\noff Tor toward non-anonymous access.\n\n2. (High) Web font served from fonts.gstatic.com\n\nIndependently of reCAPTCHA, the UI fetches a Roboto .woff2 from\nfonts.gstatic.com - a second, separate disclosure of the reporter\u0027s IP to\nGoogle on the same pages.\n\n3. (Medium) Missing Content-Security-Policy and Referrer-Policy\n\nNone of the captured first-party responses carry a Content-Security-Policy or a\nReferrer-Policy. The absence of a restrictive CSP is what permits the\ncross-origin Google/gstatic loads in findings 1\u20132; a correct CSP would have\nblocked them.\n\nObserved first-party response headers (representative):\n\nstrict-transport-security: max-age=31536000; includeSubDomains # present\nx-frame-options: SAMEORIGIN # present\nx-content-type-options: nosniff # present\naccess-control-allow-origin: https:// # scoped\n(no content-security-policy) # MISSING\n(no referrer-policy) # MISSING\n\nPoC\n\nSteps to reproduce\n\nFrom a HAR capture (or live DevTools \u2192 Network) of the submission flow:\n\n# Confirm the gate is mandatory and server-validated - the submission body\n# contains a Google reCAPTCHA token:\npython3 - \u003c\u003c\u0027PY\u0027\nimport json\nhar = json.load(open(\".har\"))\nfor e in har[\"log\"][\"entries\"]:\nif \"/whistleblower/tip?\" in e[\"request\"][\"url\"]:\nb = json.loads(e[\"request\"][\"postData\"][\"text\"])\nprint(\"captcha token present:\", bool(b.get(\"captcha\")))\nPY\n\n# Confirm the site identity is leaked to Google (anchor \"co\" param):\npython3 - \u003c\u003c\u0027PY\u0027\nimport base64, urllib.parse as u, json\nhar = json.load(open(\".har\"))\nfor e in har[\"log\"][\"entries\"]:\nq = u.parse_qs(u.urlparse(e[\"request\"][\"url\"]).query)\nif \"co\" in q:\nprint(base64.b64decode(q[\"co\"][0] + \"==\").decode(\"utf-8\", \"replace\"))\nPY\n\nExpected result\n\nAn anonymous reporting flow must complete without the reporter\u0027s browser\ncontacting any third party. The CAPTCHA must be an offline, first-party\nchallenge;\nall fonts/assets must be first-party; a strict CSP must prevent any cross-origin\nscript/font load.\n\nActual result\n\nThe submission cannot complete without a Google reCAPTCHA token; obtaining it\nrequires live connections to www.google.com/www.gstatic.com, and a web font is\nfetched from fonts.gstatic.com. Google receives the reporter\u0027s IP, a device\nfingerprint, and the exact site origin (https://:443). No\nCSP/Referrer-Policy is present to prevent this.\n\nImpact\n\nA whistleblowing platform\u0027s central promise is that a source cannot be\nidentified\nby any third party. This deployment forces every prospective source to disclose\ntheir IP address, a browser/device fingerprint, and the precise identity of the\nchannel they are using to Google - as a mandatory precondition of filing a\nreport. An adversary with visibility into, or legal reach over, that third party\ncan correlate \"who contacted reCAPTCHA for this specific whistleblowing origin,\nfrom which IP, at what time\" with the subsequently received report. The\nTor-hostility of reCAPTCHA compounds this.\n\nPotential impact:\n\n- Confidentiality: High (source deanonymization via third-party IP/fingerprint +\nsite-identity disclosure on the anonymous intake path).\n- Integrity: None.\n- Availability: Low (reCAPTCHA/Tor friction can block legitimate anonymous\nsubmissions).\n\nRemediation\n\n1. Disable Google reCAPTCHA; use an offline, first-party CAPTCHA. The intake\nflow must never depend on a third party.\n2. Self-host all fonts and assets; remove fonts.gstatic.com (and any other\nthird-party origin) from the served pages.\n3. Deploy a strict Content-Security-Policy (default-src \u0027self\u0027, no\nthird-party script-src/font-src/connect-src) and a hardened\nReferrer-Policy (e.g. no-referrer), so third-party loads are structurally\nimpossible.\n\nSeverity\n\nCritical. The defects require no privileges and no user interaction beyond a\nvictim attempting to use the platform for its intended purpose, are exploitable\nby\na network/third-party observer, and defeat the platform\u0027s primary security\nproperty (source anonymity) on the anonymous intake path (Scope: Changed).\n\nVector string\n\nCVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:N/A:L\n\nWeaknesses\n\nCommon Weakness Enumerator CWE:\n\n- CWE-359 - Exposure of Private Personal Information to an Unauthorized Actor\n- CWE-829 - Inclusion of Functionality from Untrusted Control Sphere\n- CWE-693 - Protection Mechanism Failure\n- Missing Content-Security-Policy enabling the third-party loads\n_______________________________________________\nSent through the Full Disclosure mailing list\nhttps://nmap.org/mailman/listinfo/fulldisclosure\nWeb Archives \u0026 RSS: https://seclists.org/fulldisclosure/"
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-359",
"description": "CWE-359",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-693",
"description": "CWE-693",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-829",
"description": "CWE-829",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-09-11T11:35:28Z",
"orgId": "4e2abfbf-4a2a-4b76-a4e0-d77c18ba156c",
"shortName": "VULNARCHIVE"
},
"references": [
{
"tags": [
"technical-description",
"exploit"
],
"url": "https://vuln.freearchive.org/archive/full-disclosure/2026/Jul/15"
},
{
"tags": [
"technical-description"
],
"url": "https://seclists.org/fulldisclosure/2026/Jul/15"
},
{
"url": "https://:443"
},
{
"url": "https://nmap.org/mailman/listinfo/fulldisclosure"
},
{
"url": "https://seclists.org/fulldisclosure/"
}
],
"source": {
"defect": [
"https://seclists.org/fulldisclosure/2026/Jul/15"
],
"discovery": "EXTERNAL"
},
"title": "OpenBlow Multiple Deanonymization Vulnerabilities",
"x_gcve": [
{
"recordType": "advisory",
"relationships": [],
"vulnId": "GCVE-1988-2026-0099",
"x_vulnarchive": {
"archiveUrl": "https://vuln.freearchive.org/archive/full-disclosure/2026/Jul/15",
"automated": true,
"contentSha256": "744175a5c223244ab1e223a2db5ad49da8d33cfee9f3f874cb3d0ccc5f637505",
"evidenceScore": 11,
"messageId": "",
"originalUrl": "https://seclists.org/fulldisclosure/2026/Jul/15",
"policy": "vulnarchive-1",
"sourceFormat": "text/html",
"sourcePublishedAt": "2026-06-21T10:16:53Z"
}
}
]
}
},
"cveMetadata": {
"assignerOrgId": "4e2abfbf-4a2a-4b76-a4e0-d77c18ba156c",
"assignerShortName": "VULNARCHIVE",
"datePublished": "2026-09-07T13:20:22Z",
"dateUpdated": "2026-09-11T11:35:28Z",
"state": "PUBLISHED",
"vulnId": "GCVE-1988-2026-0099"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.