FKIE_CVE-2026-107685
Vulnerability from fkie_nvd - Published: 2026-10-23 12:18 - Updated: 2026-11-02 16:17
Severity
Summary
Improper Certificate Validation vulnerability in the WebSocket streaming client in agentjido req_llm allows an on-path attacker to read and modify OpenAI WebSocket sessions, including the authorization bearer token, via a TLS certificate the client does not trust. ReqLLM.Streaming.WebSocketSession.Client.start/3 in lib/req_llm/streaming/web_socket_session/client.ex starts WebSockex without TLS options, and WebSockex defaults to insecure: true, which connects with verify: :verify_none. The option allowlist also drops any caller-supplied cacerts or ssl_options, so an application cannot turn verification on.
The OpenAI Responses, Realtime, and Codex WebSocket paths reach this client and send the API key or OAuth access token in the authorization header. HTTP streaming through Finch and the HTTP/2 duplex session through Mint verify certificates and are not affected.
This issue affects req_llm: from 1.9.0 before 1.29.0.
References
Impacted products
| Vendor | Product | Version |
|---|
{
"affected": [
{
"affectedData": [
{
"collectionURL": "https://repo.hex.pm",
"cpes": [
"cpe:2.3:a:agentjido:req_llm:*:*:*:*:*:*:*:*"
],
"defaultStatus": "unaffected",
"modules": [
"\u0027Elixir.ReqLLM.Streaming.WebSocketSession.Client\u0027",
"\u0027Elixir.ReqLLM.Streaming.WebSocketSession\u0027",
"\u0027Elixir.ReqLLM.Providers.OpenAI.WebSocket\u0027",
"\u0027Elixir.ReqLLM.OpenAI.Realtime\u0027",
"\u0027Elixir.ReqLLM.Providers.OpenAICodex\u0027"
],
"packageName": "req_llm",
"packageURL": "pkg:hex/req_llm",
"product": "req_llm",
"programFiles": [
"lib/req_llm/streaming/web_socket_session/client.ex",
"lib/req_llm/streaming/web_socket_session.ex",
"lib/req_llm/providers/openai/web_socket.ex",
"lib/req_llm/openai/realtime.ex",
"lib/req_llm/providers/openai_codex.ex"
],
"programRoutines": [
{
"name": "\u0027Elixir.ReqLLM.Streaming.WebSocketSession.Client\u0027:start/3"
},
{
"name": "\u0027Elixir.ReqLLM.Streaming.WebSocketSession\u0027:start_link/2"
},
{
"name": "\u0027Elixir.ReqLLM.Streaming.WebSocketSession\u0027:init/1"
},
{
"name": "\u0027Elixir.ReqLLM.Providers.OpenAI.WebSocket\u0027:start_responses_session/2"
},
{
"name": "\u0027Elixir.ReqLLM.OpenAI.Realtime\u0027:connect/2"
},
{
"name": "\u0027Elixir.ReqLLM.Providers.OpenAICodex\u0027:start_responses_session/2"
}
],
"repo": "https://github.com/agentjido/req_llm",
"vendor": "agentjido",
"versions": [
{
"lessThan": "1.29.0",
"status": "affected",
"version": "1.9.0",
"versionType": "semver"
}
]
},
{
"collectionURL": "https://github.com",
"cpes": [
"cpe:2.3:a:agentjido:req_llm:*:*:*:*:*:*:*:*"
],
"defaultStatus": "unaffected",
"modules": [
"\u0027Elixir.ReqLLM.Streaming.WebSocketSession.Client\u0027",
"\u0027Elixir.ReqLLM.Streaming.WebSocketSession\u0027",
"\u0027Elixir.ReqLLM.Providers.OpenAI.WebSocket\u0027",
"\u0027Elixir.ReqLLM.OpenAI.Realtime\u0027",
"\u0027Elixir.ReqLLM.Providers.OpenAICodex\u0027"
],
"packageName": "agentjido/req_llm",
"packageURL": "pkg:github/agentjido/req_llm",
"product": "req_llm",
"programFiles": [
"lib/req_llm/streaming/web_socket_session/client.ex",
"lib/req_llm/streaming/web_socket_session.ex",
"lib/req_llm/providers/openai/web_socket.ex",
"lib/req_llm/openai/realtime.ex",
"lib/req_llm/providers/openai_codex.ex"
],
"programRoutines": [
{
"name": "\u0027Elixir.ReqLLM.Streaming.WebSocketSession.Client\u0027:start/3"
},
{
"name": "\u0027Elixir.ReqLLM.Streaming.WebSocketSession\u0027:start_link/2"
},
{
"name": "\u0027Elixir.ReqLLM.Streaming.WebSocketSession\u0027:init/1"
},
{
"name": "\u0027Elixir.ReqLLM.Providers.OpenAI.WebSocket\u0027:start_responses_session/2"
},
{
"name": "\u0027Elixir.ReqLLM.OpenAI.Realtime\u0027:connect/2"
},
{
"name": "\u0027Elixir.ReqLLM.Providers.OpenAICodex\u0027:start_responses_session/2"
}
],
"repo": "https://github.com/agentjido/req_llm",
"vendor": "agentjido",
"versions": [
{
"lessThan": "c7264fc0217fbacfb5562234ff790b8e4462de76",
"status": "affected",
"version": "a389d718640ceb2d541a8f540ab954cbf5ef8313",
"versionType": "git"
}
]
}
],
"source": "6b3ad84c-e1a6-4bf7-a703-f496b71e49db"
}
],
"cveTags": [],
"descriptions": [
{
"lang": "en",
"value": "Improper Certificate Validation vulnerability in the WebSocket streaming client in agentjido req_llm allows an on-path attacker to read and modify OpenAI WebSocket sessions, including the authorization bearer token, via a TLS certificate the client does not trust. ReqLLM.Streaming.WebSocketSession.Client.start/3 in lib/req_llm/streaming/web_socket_session/client.ex starts WebSockex without TLS options, and WebSockex defaults to insecure: true, which connects with verify: :verify_none. The option allowlist also drops any caller-supplied cacerts or ssl_options, so an application cannot turn verification on.\n\nThe OpenAI Responses, Realtime, and Codex WebSocket paths reach this client and send the API key or OAuth access token in the authorization header. HTTP streaming through Finch and the HTTP/2 duplex session through Mint verify certificates and are not affected.\n\nThis issue affects req_llm: from 1.9.0 before 1.29.0."
}
],
"id": "CVE-2026-107685",
"lastModified": "2026-11-02T16:17:26.340",
"metrics": {
"cvssMetricV40": [
{
"cvssData": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "PRESENT",
"attackVector": "NETWORK",
"availabilityRequirement": "NOT_DEFINED",
"baseScore": 9.1,
"baseSeverity": "CRITICAL",
"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": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/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": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"source": "6b3ad84c-e1a6-4bf7-a703-f496b71e49db",
"type": "Secondary"
}
],
"ssvcV203": [
{
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"ssvcData": {
"id": "CVE-2026-107685",
"options": [
{
"exploitation": "poc"
},
{
"automatable": "no"
},
{
"technicalImpact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-11-02T15:22:34.733406Z",
"version": "2.0.3"
}
}
]
},
"published": "2026-10-23T12:18:06.410",
"references": [
{
"source": "6b3ad84c-e1a6-4bf7-a703-f496b71e49db",
"url": "https://cna.erlef.org/cves/CVE-2026-107685.html"
},
{
"source": "6b3ad84c-e1a6-4bf7-a703-f496b71e49db",
"url": "https://github.com/agentjido/req_llm/commit/a389d718640ceb2d541a8f540ab954cbf5ef8313"
},
{
"source": "6b3ad84c-e1a6-4bf7-a703-f496b71e49db",
"url": "https://github.com/agentjido/req_llm/commit/c7264fc0217fbacfb5562234ff790b8e4462de76"
},
{
"source": "6b3ad84c-e1a6-4bf7-a703-f496b71e49db",
"url": "https://github.com/agentjido/req_llm/security/advisories/GHSA-6vww-5m28-f26v"
},
{
"source": "6b3ad84c-e1a6-4bf7-a703-f496b71e49db",
"url": "https://osv.dev/vulnerability/EEF-CVE-2026-107685"
},
{
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"url": "https://github.com/agentjido/req_llm/security/advisories/GHSA-6vww-5m28-f26v"
}
],
"sourceIdentifier": "6b3ad84c-e1a6-4bf7-a703-f496b71e49db",
"vulnStatus": "Deferred",
"weaknesses": [
{
"description": [
{
"lang": "en",
"value": "CWE-295"
}
],
"source": "6b3ad84c-e1a6-4bf7-a703-f496b71e49db",
"type": "Secondary"
}
]
}
Loading…
Loading…
Experimental. This forecast is provided for visualization only and may change without notice. Do not use it for operational decisions.
Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.
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.
Loading…
Loading…
The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the
vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base
model, served locally by ML-Gateway.
They have not been verified by an analyst and are provided for guidance only.
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.
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.
Loading…
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.
Loading…