Vulnerability feed

CVE and vendor PSIRT advisories from 13 sources, last 30 days. Updated hourly, last run Oct 26 08:05 UTC

Critical in the last 24 hours

1601 in 30 days

Newly known exploited

In the last 24 hours, CISA KEV and vendor reports

Sources checked

Every hour, see the list

New in the last 24 hours

181-70% vs 7-day avg of 598

2 critical, 65 high, 73 medium, 31 low

Expand	Severity	Score	Advisory	Title	Source	Flags	Published
	Critical	9.1	CVE-2026-107685	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,... agentjido, OpenAIagentjido req_llm	NVD		Oct 23 12:18 UTC
CVE-2026-107685Improper Certificate Validation vulnerability in the WebSocket streaming client in agentjido req_llm allows an on-path attacker to read and modify OpenAI WebSocket sessions,... Executive summary A flaw in the req_llm Elixir library's WebSocket streaming client lets an attacker on the network path read and change OpenAI WebSocket sessions, including API keys and OAuth tokens. The advisory rates it CRITICAL with a score of 9.1. The fix is to upgrade req_llm to version 1.29.0 or later, and the advisory says the affected range is versions 1.9.0 before 1.29.0. Technical summary The weakness is improper certificate validation (CWE-295): the WebSocket client starts WebSockex without TLS options, so WebWebSockex defaults to insecure mode and skips certificate verification. An on-path attacker presenting any certificate can capture the authorization bearer token and session frames, and alter frames in both directions. The OpenAI Responses, Realtime, and Codex WebSocket paths are affected, while HTTP streaming through Finch and the HTTP/2 session through Mint are stated to verify certificates. The affected range is 1.9.0 before 1.29.0, with a git range from a389d71 before c7264fc, and the advisory states 1.29.0 returns a TLS unknown_ca error in its proof of concept. Workarounds are to keep the openai_stream_transport option at its default :sse and to avoid calling start_responses_session/2 directly; the advisory says ReqLLM.OpenAI.Realtime.connect/2 has no alternative and should not be used on affected versions. Detection guidance is not stated. Summaries are generated by AI from the linked advisory text. Confirm versions and fixes at the source. Summarized from cna.erlef.org/cves/CVE-2026-107685.html Impact if exploited ConfidentialityHigh IntegrityHigh AvailabilityNone How it is exploited Attack vector Network Attack complexity Low Attack requirements Present Privileges required None User interaction None 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 Affects agentjido OpenAI agentjido req_llm NIST National Vulnerability DatabasePublished Oct 23, 2026 12:18 UTC Read the advisory

Example	Finds
	Every word appears somewhere: title, summaries, IDs, vendor, or product
	An exact phrase
	One vendor. product: for a product, vp: for either
	Commas mean any of them, in every field
	Severity: critical, high, medium, low, or unknown
	CVSS score, with >, >=, <, or <=
	A score range
	Known exploited. is:kev works too, and kev:no leaves them out
	One source, by name
	Confidentiality impact: high, low, none, or any (low or high)
	Integrity and availability work the same way
	An advisory or CVE ID, or part of one
	A minus sign leaves out whatever the term matches

Vulnerability feed

Executive summary

Technical summary

Impact if exploited

How it is exploited

Affects

Search syntax

Welcome to CVEsee.org

What is CVEsee.org?

Disclaimer

You are leaving CVEsee.org

Install CVEsee