CVE-2026-133853 (GCVE-0-2026-133853)
Vulnerability from cvelistv5 – Published: 2026-11-20 14:44 – Updated: 2026-11-20 16:33| URL | Tags |
|---|---|
| https://github.com/flowintel/flowintel/commit/9ee… | patch |
qwen3.8:27b
2 events
advisory
bcp-05-x-01bcp-05-x-02bcp-05-x-03
Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.
| Model | Source | Identifier |
|---|---|---|
| qwen3.8:27b | ollama | qwen3.8:27b |
- Generator
-
patch2vuln.pyon 2026-11-20 14:41 - Model
qwen3.8:27b- Input
-
https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8
b2eb7327abee… - Confidence
- medium
| Commit | Subject | Patch SHA-256 |
|---|---|---|
9eec395cc73c
|
fix: [security] Isolate chatbot MCP API identities | b2eb7327abee… |
Fix summary
The shared persistent MCP session and process-wide API key mutation were removed. Each chatbot tool invocation now spawns its own short-lived MCP child process with credentials resolved fresh from the authenticated request context. If the user's identity or API key is unavailable, the invocation is rejected before any child process is launched, preventing fallback to ambient or stale credentials. Each transport is deterministically closed on success, failure, or timeout, eliminating residual credential exposure.
Patch summary
Removed the global _mcp_session, _mcp_loop, _mcp_thread, _mcp_current_api_key state and the _ensure_mcp() / _start_mcp_background() / _run_mcp_session() functions that maintained a single shared asyncio event loop and thread. Replaced with _mcp_environment() which resolves the user's API key from the authenticated session on every call and raises RuntimeError if unavailable, and _call_mcp_tool() which opens a per-invocation stdio transport, executes the tool, and closes the transport in all exit paths. Eliminated all os.environ mutations of FLOWINTEL_API_KEY. Added 33 regression tests in tests/chatbot/test_api_identity.py covering concurrent identity isolation, ambient-credential rejection, detached-user handling, key rotation, transport cleanup on failure/timeout, and real stdio child-process lifecycle.
CVSS rationale
AV:N: the chatbot is a network-accessible Flask endpoint. AC:H: exploitation requires concurrent or carefully timed requests from two different users to trigger the shared-session race, or a specific sequence of credential writes to the process environment. AT:N: no manipulation of a separate target is needed. PR:L: the attacker must be an authenticated user with chatbot access. UI:N: no victim interaction beyond their own request is required. VC:H: a user can read another user's sensitive data (private cases, investigations). VI:N, VA:N: no integrity or availability impact is evidenced. SC:N, SI:N, SA:N: no impact on separate components is demonstrated.
Weakness rationale
- CWE-290 The shared MCP session and process-wide environment variable allowed one user's authenticated identity (API key) to be applied to another user's tool invocation. The per-request identity was not isolated, effectively allowing identity confusion or spoofing between concurrent sessions.
- CWE-200 Because credentials were shared across requests and written to the process environment, a user could access data (e.g., private cases) belonging to another user whose key was cached in the shared session or ambient environment.
Attack pattern rationale
- CAPEC-126 The closest available CAPEC. The shared MCP session acted as a single long-lived session whose credentials could be 'hijacked' by a concurrent request from a different user, causing the wrong identity to be used. No CAPEC specifically covers 'shared per-tenant credential state in a multi-user service,' so CAPEC-126 is the best approximation. Uncertainty: the attack does not involve a traditional network-level session token theft but rather a server-side shared-state race.
Assumptions to verify
- AC:H is assumed because reliable exploitation requires concurrent requests from two different authenticated users or a specific timing window around the shared session restart; a single sequential request may not trigger the identity confusion.
- PR:L assumes any authenticated user with chatbot access can trigger the issue; the patch does not evidence a higher privilege requirement.
- VC:H assumes the data accessed (private cases, investigations) constitutes high confidentiality impact; the patch tests reference 'PRIVATE ADMIN INVESTIGATION' as the sensitive resource.
- CAPEC-126 (Session Hijacking) is the closest available mapping but does not precisely describe a server-side shared-credential race; no CAPEC specifically covers multi-tenant credential isolation failures in a shared in-process session.
- The affected version range is taken from the explicit analyst context (affected_version: 0, fixed_version: 4.1.0); the patch itself does not encode version boundaries.
- No evidence of active exploitation, public PoC, or in-the-wild abuse is present in the patch or metadata.
Model comparison
Selected qwen3.8:27b
by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.
| Model | Score | Agreement | Confidence | Assumptions |
|---|---|---|---|---|
qwen3.8:27b |
6 | 11 | medium | 6 |
- 2026-11-17 10:37 UTC Fix developed Corrective change authored (9eec395cc73c6706362b2c22f1e264406eb8caf8): fix: [security] Isolate chatbot MCP API identities https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8
- 2026-11-20 14:41 UTC Fix available Patch retrieved from the supplied URL; retrieval time is used as the default availability time. https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-133853",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-11-20T16:21:43.134744Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-11-20T16:33:24.293Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"cpes": [
"cpe:2.3:a:flowintel:flowintel:*:*:*:*:*:*:*:*"
],
"modules": [
"app/chatbot/chatbot_core.py"
],
"product": "flowintel",
"programFiles": [
"app/chatbot/chatbot_core.py"
],
"repo": "https://github.com/flowintel/flowintel",
"vendor": "flowintel",
"versions": [
{
"lessThan": "4.1.0",
"status": "affected",
"version": "0",
"versionType": "semver"
}
]
}
],
"cpeApplicability": [
{
"nodes": [
{
"cpeMatch": [
{
"criteria": "cpe:2.3:a:flowintel:flowintel:*:*:*:*:*:*:*:*",
"vulnerable": true
}
],
"negate": false,
"operator": "OR"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "remediation developer",
"value": "David Cruciani"
}
],
"descriptions": [
{
"lang": "en",
"value": "FlowIntel\u0027s chatbot component used a single shared MCP (Model Context Protocol) session and process-wide environment variable to pass per-user API credentials to a child process. Because the session and the ambient FLOWINTEL_API_KEY environment variable were shared across all concurrent requests, a user\u0027s authenticated identity could be applied to another user\u0027s tool invocation, or a stale/ambient credential could be inherited by a new request.\n\nPreconditions:\n- An authenticated user with access to the chatbot interface.\n- Concurrent or overlapping chatbot tool invocations from different users, or a request made after another user\u0027s credentials were written to the process environment.\n\nImpact:\n- A lower-privilege user could read data belonging to a higher-privilege user (e.g., private investigation cases).\n- A user whose session was invalidated (deleted, password changed, key revoked) could still operate under a previously cached ambient credential.\n- The parent process environment was mutated with user-specific secrets, creating a persistent credential leak surface.\n\nAffected versions: FlowIntel prior to 4.1.0."
}
],
"impacts": [
{
"capecId": "CAPEC-126",
"descriptions": [
{
"lang": "en",
"value": "CAPEC-126 Session Hijacking"
}
]
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "HIGH",
"attackRequirements": "NONE",
"attackVector": "NETWORK",
"baseScore": 6,
"baseSeverity": "MEDIUM",
"privilegesRequired": "LOW",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:H/AT:N/PR:L/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "NONE",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "NONE",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"format": "SSVC",
"other": {
"content": {
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "Supplier",
"timestamp": "2026-11-20T14:41:59Z",
"version": "2.0.3"
},
"type": "SSVC"
},
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-290",
"description": "CWE-290 Authentication Bypass by Spoofing",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-200",
"description": "CWE-200 Exposure of Sensitive Information to an Unauthorized Actor",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-11-20T14:44:31.341Z",
"orgId": "5a6e4751-2f3f-4070-9419-94fb35b644e8",
"shortName": "CIRCL"
},
"references": [
{
"name": "Security patch",
"tags": [
"patch"
],
"url": "https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8"
}
],
"solutions": [
{
"lang": "en",
"value": "The shared persistent MCP session and process-wide API key mutation were removed. Each chatbot tool invocation now spawns its own short-lived MCP child process with credentials resolved fresh from the authenticated request context. If the user\u0027s identity or API key is unavailable, the invocation is rejected before any child process is launched, preventing fallback to ambient or stale credentials. Each transport is deterministically closed on success, failure, or timeout, eliminating residual credential exposure."
}
],
"title": "FlowIntel Chatbot Shared MCP Session Allows Cross-User API Identity Confusion",
"x_gcve": [
{
"extensions": {
"bcp-05-x-01": {
"ai_annotations": [
{
"ai_level": "generated",
"description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.",
"gna_source": 1,
"models": [
{
"gna_source": 1,
"identifier": "qwen3.8:27b",
"name": "qwen3.8:27b",
"source": "ollama"
}
],
"review_status": "full",
"scope": "record",
"tags": [
"ai-computer-assisted:llm-generated",
"ai-computer-assisted:classification"
]
}
]
},
"bcp-05-x-02": {
"x_patch2vuln": {
"assumptions": [
"AC:H is assumed because reliable exploitation requires concurrent requests from two different authenticated users or a specific timing window around the shared session restart; a single sequential request may not trigger the identity confusion.",
"PR:L assumes any authenticated user with chatbot access can trigger the issue; the patch does not evidence a higher privilege requirement.",
"VC:H assumes the data accessed (private cases, investigations) constitutes high confidentiality impact; the patch tests reference \u0027PRIVATE ADMIN INVESTIGATION\u0027 as the sensitive resource.",
"CAPEC-126 (Session Hijacking) is the closest available mapping but does not precisely describe a server-side shared-credential race; no CAPEC specifically covers multi-tenant credential isolation failures in a shared in-process session.",
"The affected version range is taken from the explicit analyst context (affected_version: 0, fixed_version: 4.1.0); the patch itself does not encode version boundaries.",
"No evidence of active exploitation, public PoC, or in-the-wild abuse is present in the patch or metadata."
],
"capecRationale": [
{
"capecId": "CAPEC-126",
"rationale": "The closest available CAPEC. The shared MCP session acted as a single long-lived session whose credentials could be \u0027hijacked\u0027 by a concurrent request from a different user, causing the wrong identity to be used. No CAPEC specifically covers \u0027shared per-tenant credential state in a multi-user service,\u0027 so CAPEC-126 is the best approximation. Uncertainty: the attack does not involve a traditional network-level session token theft but rather a server-side shared-state race."
}
],
"commit": "9eec395cc73c6706362b2c22f1e264406eb8caf8",
"confidence": "medium",
"credits": [
{
"lang": "en",
"type": "remediation developer",
"value": "David Cruciani"
}
],
"cvssRationale": "AV:N: the chatbot is a network-accessible Flask endpoint. AC:H: exploitation requires concurrent or carefully timed requests from two different users to trigger the shared-session race, or a specific sequence of credential writes to the process environment. AT:N: no manipulation of a separate target is needed. PR:L: the attacker must be an authenticated user with chatbot access. UI:N: no victim interaction beyond their own request is required. VC:H: a user can read another user\u0027s sensitive data (private cases, investigations). VI:N, VA:N: no integrity or availability impact is evidenced. SC:N, SI:N, SA:N: no impact on separate components is demonstrated.",
"fixSummary": "The shared persistent MCP session and process-wide API key mutation were removed. Each chatbot tool invocation now spawns its own short-lived MCP child process with credentials resolved fresh from the authenticated request context. If the user\u0027s identity or API key is unavailable, the invocation is rejected before any child process is launched, preventing fallback to ambient or stale credentials. Each transport is deterministically closed on success, failure, or timeout, eliminating residual credential exposure.",
"generatedAt": "2026-11-20T14:41:59.093131Z",
"generator": "patch2vuln.py",
"model": "qwen3.8:27b",
"modelComparison": {
"potentialErrors": [],
"rankings": [
{
"agreementScore": 11,
"assumptionCount": 6,
"confidence": "medium",
"model": "qwen3.8:27b",
"score": 6
}
],
"selectedModel": "qwen3.8:27b",
"selectionMethod": "deterministic-consensus-v1",
"selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
},
"patchSha256": "b2eb7327abee28d605ef02f5af16f090ce7a643bd63bf2a5836e30539adfbb81",
"patchSummary": "Removed the global _mcp_session, _mcp_loop, _mcp_thread, _mcp_current_api_key state and the _ensure_mcp() / _start_mcp_background() / _run_mcp_session() functions that maintained a single shared asyncio event loop and thread. Replaced with _mcp_environment() which resolves the user\u0027s API key from the authenticated session on every call and raises RuntimeError if unavailable, and _call_mcp_tool() which opens a per-invocation stdio transport, executes the tool, and closes the transport in all exit paths. Eliminated all os.environ mutations of FLOWINTEL_API_KEY. Added 33 regression tests in tests/chatbot/test_api_identity.py covering concurrent identity isolation, ambient-credential rejection, detached-user handling, key rotation, transport cleanup on failure/timeout, and real stdio child-process lifecycle.",
"patchTruncated": false,
"patches": [
{
"commit": "9eec395cc73c6706362b2c22f1e264406eb8caf8",
"date": "Tue, 17 Nov 2026 11:37:34 +0100",
"patchSha256": "b2eb7327abee28d605ef02f5af16f090ce7a643bd63bf2a5836e30539adfbb81",
"retrievedAt": "2026-11-20T14:41:10.165065Z",
"source": "https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8",
"sourceUrl": "https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8",
"subject": "fix: [security] Isolate chatbot MCP API identities"
}
],
"source": "https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8",
"ssvc": {
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "Supplier",
"timestamp": "2026-11-20T14:41:59Z",
"version": "2.0.3"
},
"subject": "fix: [security] Isolate chatbot MCP API identities",
"weaknessRationale": [
{
"cweId": "CWE-290",
"rationale": "The shared MCP session and process-wide environment variable allowed one user\u0027s authenticated identity (API key) to be applied to another user\u0027s tool invocation. The per-request identity was not isolated, effectively allowing identity confusion or spoofing between concurrent sessions."
},
{
"cweId": "CWE-200",
"rationale": "Because credentials were shared across requests and written to the process environment, a user could access data (e.g., private cases) belonging to another user whose key was cached in the shared session or ambient environment."
}
]
}
},
"bcp-05-x-03": {
"x_timeline": {
"events": [
{
"description": "Corrective change authored (9eec395cc73c6706362b2c22f1e264406eb8caf8): fix: [security] Isolate chatbot MCP API identities",
"id": "evt-fix-developed-1",
"references": [
"https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8"
],
"timestamp": "2026-11-17T10:37:34Z",
"type": "fix-developed"
},
{
"description": "Patch retrieved from the supplied URL; retrieval time is used as the default availability time.",
"id": "evt-fix-available-1",
"references": [
"https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8"
],
"timestamp": "2026-11-20T14:41:10.165065Z",
"type": "fix-available"
}
]
}
}
},
"recordType": "advisory",
"vulnId": "GCVE-1-2026-20421"
}
]
}
},
"cveMetadata": {
"assignerOrgId": "5a6e4751-2f3f-4070-9419-94fb35b644e8",
"assignerShortName": "CIRCL",
"cveId": "CVE-2026-133853",
"datePublished": "2026-11-20T14:44:31.341Z",
"dateReserved": "2026-11-20T14:44:25.304Z",
"dateUpdated": "2026-11-20T16:33:24.293Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2",
"vulnerability-lookup:meta": {
"nvd": {
"cve": {
"affected": [
{
"affectedData": [
{
"cpes": [
"cpe:2.3:a:flowintel:flowintel:*:*:*:*:*:*:*:*"
],
"modules": [
"app/chatbot/chatbot_core.py"
],
"product": "flowintel",
"programFiles": [
"app/chatbot/chatbot_core.py"
],
"repo": "https://github.com/flowintel/flowintel",
"vendor": "flowintel",
"versions": [
{
"lessThan": "4.1.0",
"status": "affected",
"version": "0",
"versionType": "semver"
}
]
}
],
"source": "5a6e4751-2f3f-4070-9419-94fb35b644e8"
}
],
"cveTags": [],
"descriptions": [
{
"lang": "en",
"value": "FlowIntel\u0027s chatbot component used a single shared MCP (Model Context Protocol) session and process-wide environment variable to pass per-user API credentials to a child process. Because the session and the ambient FLOWINTEL_API_KEY environment variable were shared across all concurrent requests, a user\u0027s authenticated identity could be applied to another user\u0027s tool invocation, or a stale/ambient credential could be inherited by a new request.\n\nPreconditions:\n- An authenticated user with access to the chatbot interface.\n- Concurrent or overlapping chatbot tool invocations from different users, or a request made after another user\u0027s credentials were written to the process environment.\n\nImpact:\n- A lower-privilege user could read data belonging to a higher-privilege user (e.g., private investigation cases).\n- A user whose session was invalidated (deleted, password changed, key revoked) could still operate under a previously cached ambient credential.\n- The parent process environment was mutated with user-specific secrets, creating a persistent credential leak surface.\n\nAffected versions: FlowIntel prior to 4.1.0."
}
],
"id": "CVE-2026-133853",
"lastModified": "2026-11-20T17:18:23.500",
"metrics": {
"cvssMetricV40": [
{
"cvssData": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "HIGH",
"attackRequirements": "NONE",
"attackVector": "NETWORK",
"availabilityRequirement": "NOT_DEFINED",
"baseScore": 6.0,
"baseSeverity": "MEDIUM",
"confidentialityRequirement": "NOT_DEFINED",
"exploitMaturity": "NOT_DEFINED",
"integrityRequirement": "NOT_DEFINED",
"modifiedAttackComplexity": "NOT_DEFINED",
"modifiedAttackRequirements": "NOT_DEFINED",
"modifiedAttackVector": "NOT_DEFINED",
"modifiedPrivilegesRequired": "NOT_DEFINED",
"modifiedSubAvailabilityImpact": "NOT_DEFINED",
"modifiedSubConfidentialityImpact": "NOT_DEFINED",
"modifiedSubIntegrityImpact": "NOT_DEFINED",
"modifiedUserInteraction": "NOT_DEFINED",
"modifiedVulnAvailabilityImpact": "NOT_DEFINED",
"modifiedVulnConfidentialityImpact": "NOT_DEFINED",
"modifiedVulnIntegrityImpact": "NOT_DEFINED",
"privilegesRequired": "LOW",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:H/AT:N/PR:L/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
"version": "4.0",
"vulnAvailabilityImpact": "NONE",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "NONE",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"source": "5a6e4751-2f3f-4070-9419-94fb35b644e8",
"type": "Secondary"
}
],
"ssvcV203": [
{
"source": "5a6e4751-2f3f-4070-9419-94fb35b644e8",
"ssvcData": {
"options": [
{
"exploitation": "none"
},
{
"automatable": "yes"
},
{
"technicalImpact": "partial"
}
],
"role": "Supplier",
"timestamp": "2026-11-20T14:41:59Z",
"version": "2.0.3"
}
},
{
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"ssvcData": {
"id": "CVE-2026-133853",
"options": [
{
"exploitation": "none"
},
{
"automatable": "no"
},
{
"technicalImpact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-11-20T16:21:43.134744Z",
"version": "2.0.3"
}
}
]
},
"published": "2026-11-20T15:17:53.947",
"references": [
{
"source": "5a6e4751-2f3f-4070-9419-94fb35b644e8",
"url": "https://github.com/flowintel/flowintel/commit/9eec395cc73c6706362b2c22f1e264406eb8caf8"
}
],
"sourceIdentifier": "5a6e4751-2f3f-4070-9419-94fb35b644e8",
"vulnStatus": "Received",
"weaknesses": [
{
"description": [
{
"lang": "en",
"value": "CWE-200"
},
{
"lang": "en",
"value": "CWE-290"
}
],
"source": "5a6e4751-2f3f-4070-9419-94fb35b644e8",
"type": "Secondary"
}
]
}
}
}
}
Solutions
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.