Authentication Bypass

An authentication bypass in OpenBMC's phosphor-net-ipmid IPMI stack lets a network attacker with no credentials complete IPMI 2.0 RAKP session authentication and obtain full BMC control. A crafted RAKP Message 1 forces the handler to return before the authentication object's constructor defaults are overwritten, so the service accepts a RAKP Message 3 whose HMAC is keyed with the hard-coded 20-byte 'userKey' derived from the string '0penBmc' combined with an often-predictable 'bmcRandomNum'. The flaw affects downstream vendors that ship phosphor-net-ipmid as their IPMI implementation, including NVIDIA and H3C, and runZero published a public advisory; no CISA KEV entry or EPSS score was supplied, and no patched version was identified in the provided data.

Privilege escalation in OpenBMC's phosphor-net-ipmid IPMI stack allows an attacker who already holds a valid low-privileged IPMI session to redirect that session's authorization context to a different, higher-privileged account while the original integrity and encryption keys stay in place, so no re-authentication ever occurs. Downstream vendors that package phosphor-net-ipmid — NVIDIA and H3C among them — inherit the flaw in their BMC firmware. Rated CVSS 8.8 with high confidentiality, integrity and availability impact; no public exploit code identified at time of analysis, and the issue is not listed in CISA KEV.

Cross-organization user administration takeover in Casdoor through 4.4.0 allows any holder of a single application's clientId and clientSecret to act on user accounts across all organizations. The flaw is an authorization bypass in the /api/mcp endpoint that exposes password salts, email addresses, and account creation/modification/deletion capabilities. Public exploit documentation exists, but no active exploitation status or EPSS data was provided.

Unauthenticated information disclosure in Dromara lamp-cloud through 5.10.0 stems from an overly broad anonymous-access whitelist pattern (/*/anno/**) that exposes the /defGenProject/anno/getProperties endpoint. Any remote attacker who can reach the service via HTTP can POST to that path and receive the server's complete JVM system property map, leaking classpath entries, filesystem paths, OS build details, and startup secrets injected as properties. Exploit code is publicly available (a writeup and proof-of-concept are linked from the CVE references and VulnCheck advisory), and the flaw has not been listed in CISA KEV, so it is a high-value reconnaissance and credential-harvesting primitive rather than a confirmed mass-exploitation event.

We need answer JSON only. Need analyze CVE. Need fill each field string. Need use all data. Need be careful. We need produce JSON object valid. Let's think. CVE-2026-91995. Description: pig before 4.1.0 authentication bypass in /register/password endpoint where password verification results discarded, allowing any value as current password. Remote attackers can submit username with incorrect current password to overwrite any account credential including admin gain full admin control. Product: pig? pig-mesh/pig. Need product_name 1-3 words: "Pig" maybe "Pig Mesh". Specific product/service. Affected: pig-mesh:pig. Likely "Pig" or "Pig Mesh". Pick "Pig" maybe primary. But product_name should be specific. "Pig" is open-source microservices platform? Use "Pig" maybe. Could "Pig4Cloud"? CPE vendor pig-mesh, product pig. "Pig" fine. Summary: Need original synthesis. Mention authentication bypass in Pig before 4.1.0 in /register/password allows remote unauthenticated? CVSS 4.0 PR:N. We can say remote unauthenticated attackers can call password reset/register password endpoint with arbitrary current password to reset any account, including admin. POC public. Patch. EPSS? Not provided, don't mention. KEV? Not provided. Need infer no KEV. Use "no public exploit identified"? But POC publicly available. Use "publicly available exploit code exists". Not actively exploited per no KEV. Need mention patch. Good. Technical_context: Java Spring? Need identify technology/library/protocol. CWE-620 Unverified Password Change. Endpoint /register/password in pig-upms. SysUserServiceImpl resetUserPassword calls checkPassword but ignored result; patch checks Boolean.TRUE.equals(checkedPassword.getData()). Need educational. CPE exact affected products cpe:2.3:a:pig-mesh:pig:*:*:*:*:*:*:*:* all versions before 4.1.0? CPE says * but description before 4.1.0. Explain root cause class. Risk_assessment: Use CVSS 4.0 vector AV:N AC:L AT:N PR:N UI:N VC:H VI:H VA:N SC:N SI:N SA:N => critical. E

Semaphore UI through 2.19.12 fails to enforce project-level resource permissions on GET and HEAD requests in its GetMustCanMiddleware, letting any authenticated user holding only the low-privileged guest or task_runner role read every project's environment records. Those environment records store plaintext secrets, credentials, and passwords used to reach managed infrastructure, so a read-only account can effectively harvest the keys to the entire estate. Publicly available exploit code exists (tracked by VulnCheck, issue #4150), the flaw is rated 7.1 with a network attack vector and low privileges required, and no CISA KEV listing has been observed at time of analysis.

We need to produce JSON. Let me analyze. CVE-2026-91924: pgweb through 0.17.0, missing authorization on POST /api/connect. CWE-862 (Missing Authorization). CVSS 4.0 vector: AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:L/VA:N/SC:H/SI:L/SA:N → base score roughly 8.4? Let's check: CVSS 4.0 with PR:L, VC:H, VI:L, SC:H, SI:L → probably ~8.4 is plausible? Actually with PR:L and no VA/SA... Let's just take given 8.4. PR:L means authenticated (or requires some privileges). So "authenticated" is allowed since vector contains PR:L. No KEV, no POC mentioned. So "no public exploit identified at time of analysis". Hmm, but there's a GitHub issue #869 — that's a report but not necessarily public exploit code. Patch status: no fix version given; references include a GitHub issue and code at a specific commit 6b0b0244... That commit hash could be before or after fix. Hmm. The references point to routes.go and api.go at commit 6b0b0244d1aefd6971999b03481eeeaa4ec7cf55 — presumably the vulnerable code. No fix version provided. So "No vendor-released patch identified at time of analysis" — but careful, absence of data is not confirmation. Actually pgweb is at 0.17.0 "through" which suggests the latest version is vulnerable. Rule says use most specific that applies; if patch data says NOT available → "No vendor-released patch identified at time of analysis". We don't have patch data explicitly. I'll say no vendor-released patch identified in the provided intel; issue #869 referenced. Product name: "pgweb" — 1-3 words. Just "pgweb". Summary: must not start with "A vulnerability". Start with specific impact verb and product. E.g. "Missing authorization in pgweb through 0.17.0 lets anyone able to reach the web UI's connect endpoint..." Note PR:L — requires low privileges, i.e., a session/authenticated user of pgweb UI. Write original synthesis. Technical context: pgweb is a Go-based, cross-platform PostgreSQL web client/browser (sosedoff/pgweb). Connect-backend authorization (--connect-backend?

We need answer JSON only. Need understand CVE. Need synthesize. Need fill fields. Let's parse data. CVE-2026-57147. PraisonAI multi-agent system. Prior to 0.1.6, praisonai_platform/services/auth_service.py assigns public dev-secret-change-me value to JWT_SECRET when PLATFORM_JWT_SECRET unset, production guard does not run when PLATFORM_ENV unset because defaults dev. Remote unauthenticated attacker can mint HS256 token arbitrary sub/email, AuthService._verify_token() and get_current_user dependency accept forged identity for protected API routes. Fixed in praisonai-platform 0.1.6. CVSS 9.8 AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H. CWE-798. Reported by GitHub_M. CPEs: cpe:2.3:a:mervinpraison:praisonai:*:*:*:*:*:*:*:*, cpe:2.3:a:mervinpraison:praisonai-platform:*:*:*:*:*:*:*:* Tags: Authentication Bypass, Praisonai, Praisonai Platform References: GHSA-cwj8-7gp2-ggcw, PR 1793, commit e0fb..., release tag v4.6.51 (interesting? maybe unrelated release?). Source code evidence: advisory summary. Package praisonai-platform PyPI latest tested 0.1.4, current 2026-06-01. Weakness CWE-798 + CWE-1188. Detail of code. Also fix diff shows examples eval changes, not auth_service? Maybe PR includes unrelated? But description says fixed in 0.1.6. "Package: pip/praisonai-platform — vulnerable: <= 0.1.4 — fixed in: 0.1.6" Hmm maybe <=0.1.5? We say prior to 0.1.6. Need product_name: 1-3 words. likely "PraisonAI Platform" or "praisonai-platform". Use "PraisonAI Platform". Summary not copy description. Need mention specific. Technical context. Risk assessment: CVSS 9.8, EPSS? Not provided. We need say EPSS unknown/not provided. KEV? Input no KEV field. We have no KEV status. POC? Not provided, but exploit technique described publicly in GHSA. Need use only phrases: if neither POC no KEV? But source advisory with detailed exploit maybe no explicit POC. We can say "no public exploit identified at time of analysis" perhaps but technical advisory demonstrates forging token. Need not overstate

We need answer in JSON only. Need synthesize. Need be careful with all fields. Need extract data. We have CVE-2026-57148. Description: PraisonAI multi-agent teams system. Prior to 0.1.6, praisonai_platform/services/auth_service.py falls back to public dev-secret-change-me HS256 signing key when PLATFORM_JWT_SECRET unset, while guards disabled because PLATFORM_ENV defaults to dev. Unauthenticated attacker can sign JWT with attacker-chosen sub, AuthService._verify_token accepts, enabling user/workspace-owner impersonation when target identifier known. Fixed in praisonai-platform 0.1.6. CVSS 9.8 vector AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H CWE-287. Multi-source: Reported by GitHub_M. CPE cpe:2.3:a:mervinpraison:praisonai:*:*:*:*:*:*:*:*, cpe:2.3:a:mervinpraison:praisonai-platform:*:*:*:*:*:*:*:*. Tags: Authentication Bypass, Praisonai, Praisonai Platform. References GHSA-f38v-77qj-h4jq, PR 1793, commit e0fb8e7..., release v4.6.51. Source code evidence from GHSA advisory: affected <=0.1.4 including 0.1.4 declared patch? main HEAD affected. CWE-1188 + CWE-798 -> CWE-287. Details. Package pip/praisonai-platform vulnerable <=0.1.4 fixed in 0.1.6. PR diff includes examples eval changes irrelevant? Need mention maybe commit/PR contains fix? It seems PR diff includes safe eval changes? But source code evidence says PR diff is maybe unrelated? Need synthesize. We have no EPSS, KEV, POC input. Need use only provided data. KEV not provided -> no KEV; POC? Not explicitly provided, but exploit details? The advisory describes forging token with public key, so maybe exploit technique public but no explicit POC flag. Need say no public exploit identified? But source code evidence includes detailed steps. There is no "POC" field. We can say public exploit code? Careful: "publicly available exploit code exists" if POC=YES but no KEV. Input didn't provide POC. There is no explicit POC. However advisory describes exploitation procedure, but not "POC code". We should use "no public exploi

We need answer JSON only. Need synthesize. Need obey fields. Need use boleslawiec in risk_assessment only, lowercase. Need not mention note. Let's parse input. CVE-2026-57140 Description: PraisonAI multi-agent teams. From 1.6.0 until 1.7.2, AgentOS in src/praisonai-ts/src/os/agentos.ts uses 0.0.0.0 default and registers GET /api/agents and POST /api/chat without auth middleware. Remote caller can obtain agent names, roles, instruction prefixes and invoke selected agent, potentially reaching tools, memory, external APIs, credentials, workflow state. Initial remediation released in 1.7.2. CVSS Score 9.4, Vector CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:L CWE-306 missing authentication. Reported by GitHub_M CPE: cpe:2.3:a:mervinpraison:praisonai:*:*:*:*:*:*:*:* Tags: Authentication Bypass, Praisonai References: GHSA-9752-mhqh-h34f commit releases/tag/v4.6.60 Source code evidence: GHSA summary. npm package praisonai, AgentOS HTTP server defaults 0.0.0.0 and no auth. GET /api/agents, POST /api/chat. Affected npm/praisonai >=1.6.0 <=1.7.1 fixed in 1.7.2. Distinct from Python/PyPI. Current published npm 1.7.1. Same TS source in origin/main at v4.6.58. Technical details config. Default CORS *. app listens all interfaces unless overridden. Docs. Package: npm/praisonai vulnerable >=1.6.0, <=1.7.1 fixed in 1.7.2. Commit diff shows Dockerfiles bump from praisonai>=4.6.59 to >=4.6.60, version bumps in pyproject to 1.6.60, TS changes unrelated? Actually commit link likely maybe release? It includes multiple changes. But advisory says fixed in npm 1.7.2. Also has releases/tag/v4.6.60. Hmm maybe CVE references say release v4.6.60 but package fix 1.7.2. Need be cautious: "Initial remediation was released in version 1.7.2." NVD description says fixed 1.7.2. GHSA package says fixed 1.7.2. Commit diff perhaps v4.6.60 in other package python. Need mention patch version 1.7.2 for npm. Need determine product_name: "PraisonAI" maybe 1-3 words. Could be "PraisonAI AgentOS" or "Pra

We need answer JSON only. Need synthesize. We need adhere to all fields. Let's parse input. CVE-2026-57139 Description: PraisonAI multi-agent teams system. From 1.5.0 until 1.7.2, MCPServer.startHttp() in src/praisonai-ts/src/mcp/server.ts binds without host restriction and forwards every HTTP POST request to handleRequest() without authentication or authorization. Any network client that can reach the port can call tools/list, tools/call, resources/read, or prompts/get, causing registered handlers to run with server-side credentials and process privileges or disclose registered data. An initial remediation was released in version 1.7.2. CVSS Score: 9.8 CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H CWE: CWE-306 Reported by GitHub_M CPE: cpe:2.3:a:mervinpraison:praisonai:*:*:*:*:*:*:*:* Tags: Authentication Bypass, Praisonai References: advisory GHSA, commit, release tag v4.6.60 Source evidence GHSA says package npm/praisonai vulnerable >=1.5.0 <=1.7.1 fixed 1.7.2. Also references release tag v4.6.60 maybe Python? Need note discrepancy? CVE description says initial remediation in 1.7.2. Fix version 1.7.2 for npm package. Release tag 4.6.60 is maybe different packaging. Need mention. Need product_name: "PraisonAI" likely 1 word but okay. They ask 1-3 words exact product. Could be "PraisonAI" maybe "PraisonAI MCPServer"? primary product. Use "PraisonAI". Summary: original synthesis, not copy. Start with impact verb and product. Need mention actively exploited? Input KEV absent? No KEV field. POC? no explicit POC. EPSS? not given. We can say no public exploit identified. But need not mention if missing? It asks if POC exists and EPSS risk if notable. We don't have EPSS. So "No public exploit identified at time of analysis; EPSS not provided." Need maybe avoid. "Unauthenticated remote code execution in PraisonAI's TypeScript MCPServer..." Wait CVE is missing auth leading to invoking tools, could cause RCE? CVSS C/I/A H but description says handlers run wit

Let me synthesize. Key facts: - CVE-2026-57134, PraisonAI (npm package "praisonai", TypeScript MCP security helper). - MCPSecurity.evaluatePolicy() only validates credentials for api-key/bearer; basic/oauth accept any non-empty Authorization header, skipping validate() callback, returning authenticated: true. - Affected: npm praisonai >= 1.5.1, <= 1.7.1; fixed in 1.7.2. CPE cpe:2.3:a:mervinpraison:praisonai:*:*:*:*:*:*:*:* - CVSS 8.2 AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N. But logically, exploitation requires the deployment to use basic or oauth auth policy — that's a configuration prerequisite. Hmm. CVSS PR:N given. My assessed vector: I'd argue PR:N is defensible from the attacker's perspective (attacker sends an arbitrary Authorization header without valid credentials). But the configuration requirement: the target must use basic/oauth policy. AV:N is fine. AC:L? Arguably AC:H because it requires non-default config (basic/oauth policy configured). Let me think: the guidance says choose PR honestly — if exploitation requires control over configuration files, admin access, or authentication, use PR:H or PR:L. Here the attacker doesn't need config access; they need the target to have configured basic or oauth. That's a configuration precondition, not attacker privilege. I'd keep PR:N and use AC:H? Hmm. Many would keep AC:L. Actually the exploit is trivially automatable once the target uses basic/oauth. I'll go AC:L, PR:N, but note the pre-condition. Hmm, but the rubric's example: "Requires write access to the httpd configuration, so PR:H and AC:H". That's different. Here attacker just sends header. I'll use AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N — matching but I should express independent judgment. Maybe I could argue I:L and A:N as given. Let me consider C:H — access to MCP tools and resources protected by policies → confidentiality high, integrity low. That's reasonable. I'll go with AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N — but then it's a copy. The instruction says "NOT

Unauthenticated MCP tool invocation in PraisonAI (praisonaiagents 0.6.0–1.6.58 and praisonai 3.10.0–4.6.58) lets a malicious website use DNS rebinding to reach a local or internal ToolsMCPServer SSE endpoint on /sse and /messages/, enumerate registered tools, and invoke them with the server user's privileges. The legacy SSE wrapper performs no Host, Origin, or authentication enforcement, while the newer Streamable HTTP transport rejects the same hostile Origin with HTTP 403, isolating the flaw to the legacy SSE path. An initial remediation ships in praisonaiagents

MISP through 2.5.45 fails to enforce authorization checks when collection elements are created, allowing an authenticated user to persist a bare UUID referencing an Event or Galaxy Cluster they are not permitted to read. The collection view later resolves that UUID back into the underlying object, so collections become a handle on another organisation's private data — the patch notes this already caused org-only events to be disclosed in the beta collection view. Exploitation requires a valid MISP account and knowledge of the target UUID; no public exploit code or CISA KEV listing was identified at time of analysis, though the CVSS 4.0 vector (PR:L, VC:H) and EPSS-adjacent signals mark it as a meaningful confidentiality risk for shared threat-intel instances.

An authorization bypass in MISP through version 2.5.45 lets an authenticated user who can edit events assign an arbitrary sharing_group_id they are not authorized to use, because the authorization check only ran when the request explicitly supplied distribution = 4. If the caller omits the distribution field, MISP's field-recovery logic restores the event's stored distribution (4, for sharing-group events) and persists the unauthorized sharing group ID, silently re-scoping who can see that event and its threat-intelligence content. No public exploit code or active exploitation has been identified at time of analysis, but exploitation is a single crafted edit request, so the integrity impact (VI:H, CVSS 4.0 base 7.1) is significant for multi-tenant or community-sharing MISP instances.

Incorrect authorization validation in Octopus Server permits authenticated users holding certain scoped permission sets to execute arbitrary scripts on deployment workers, including the built-in worker that runs directly within the Octopus Server process. The flaw bypasses the permission check that should gate script execution, allowing a user to operate outside their authorized scope without possessing the required authorisation. No public exploit has been identified and the CVE is not in CISA KEV, but the low authentication barrier - any user with some scoped permissions - makes this a meaningful privilege-escalation risk in CI/CD environments where Octopus Server routinely holds deployment credentials and infrastructure access.

Trust-boundary spoofing in the proxy-addr Node.js module (versions 1.1.0 through 2.0.7), the library that backs Express req.ip and req.ips, lets unauthenticated clients forge their apparent source address. When an operator declares a trusted-proxy subnet in IPv4-mapped IPv6 notation with an IPv4-sized prefix (e.g. ::ffff:10.0.0.0/8 rather than the correct /104), proxy-addr silently accepts it but trusts every IPv4 address on the internet, so the direct socket peer becomes trusted at hop 0 and any client can set an arbitrary X-Forwarded-For header. This is a fail-open regression introduced in 1.1.0, fixed in 2.0.8; no public exploit identified at time of analysis, and it is not in CISA KEV.

Insufficient session expiration in bizwell xClick groupware allows authentication bypass through replay of session tokens that should have been invalidated after logout or timeout. Affecting versions R2, R3, and R3.1, the flaw enables an attacker who has obtained a victim's session token to continue using it with full account-level access even after the legitimate session should have ended, carrying high confidentiality, integrity, and availability impact per the CVSS 4.0 score of 8.7. No public exploit code has been identified at time of analysis, and the vulnerability does not appear in the CISA Known Exploited Vulnerabilities catalog.

Authorization bypass in IceHRM before 36.0.0 lets any authenticated employee read coworkers' HR records by substituting an arbitrary employee ID in seven REST sub-resource endpoints covering skills, education, certifications, languages, leave, attendance and status. CVSS 4.0 scores the flaw 7.1 (AV:N/AC:L/PR:L/UI:N, VC:H only), reflecting a low-effort, easily automated disclosure of sensitive personnel data rather than code execution. No public exploit code was identified at time of analysis and the issue is not listed in CISA KEV; a vendor fix is available in release v36.0.0.

Unauthenticated retrieval of Stripe Checkout Session objects in the MarcosCamara01 Ecommerce Template (before commit 91e273c) exposes buyer personally identifiable information to anyone holding a session id. The GET handler at /api/stripe/checkout_sessions calls stripe.checkout.sessions.retrieve() and returns the full object — buyer name, email, phone, billing address, amount paid and internal userId — without ever calling auth.api.getSession() or verifying that the session belongs to the caller, even though sibling endpoints such as POST /api/stripe/payment already enforced authentication. Exploitation is not confirmed in the wild and no public exploit code was identified, but the session_id is written into the buyer's own success_url (/result?session_id={CHECKOUT_SESSION_ID}), so it leaks through Referer headers, analytics, access logs and shared-machine browser history, making the flaw practically reachable rather than purely theoretical.

Improper access control in IBM Sterling File Gateway (6.2.0.0 through 6.2.0.6_1, 6.2.1.0-6.2.1.2, and 6.2.2.0-6.2.2.1) lets a remote, unauthenticated attacker reach file-transfer data they should not be able to see. The CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N, C:H) describes a trivially reachable, no-credentials-required read of sensitive information with no integrity or availability impact. No public exploit code or CISA KEV entry was present in the supplied intelligence at the time of analysis, and IBM has published a vendor fix.

Local privilege escalation in Apple macOS lets an application running with standard user privileges obtain root, giving an attacker full control of the host once code executes on the machine. The flaw is an improper access-control check (CWE-284) that Apple addressed with improved validation logic, fixed in macOS Sequoia 15.8, macOS Tahoe 26.7, and macOS Golden Gate 27. No public exploit code has been identified and the vulnerability is not in CISA KEV; CISA's SSVC assessment rates exploitation as 'none' and automatable as 'no', so risk is concentrated in environments where untrusted local code can be executed.

Apple macOS contains a local privilege-escalation flaw in which an application running under an ordinary user account can abuse insufficient access-control checks to obtain root privileges. Devices are affected if they run macOS versions below Sequoia 15.8, Tahoe 26.7 or Golden Gate 27, and Apple has shipped fixes for all three release lines. CVSS 3.1 scores this 7.8 (AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H) and CISA's SSVC rates technical impact as total, but there is no CISA KEV listing, no reported exploitation and no public exploit code identified at time of analysis, so this is best treated as a high-value post-compromise escalation primitive rather than an imminent remote threat.

A local privilege-escalation flaw in Apple's operating system family lets a malicious app already running on the device obtain root privileges by exploiting a missing authorization check. Apple addressed the issue with 'additional restrictions' in iOS 26.7 and iPadOS 26.7, iOS 27 and iPadOS 27, macOS Golden Gate 27, and visionOS 27, and the CVSS 3.1 score is 7.8 (AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H). Exploitation is not confirmed: the issue is not in CISA KEV, no public exploit code is identified, and CISA's SSVC assessment lists exploitation as 'none' and automatable as 'no', though technical impact is rated 'total' once root is achieved.

Missing authorization controls in the ThemeAtelier Domain For Sale WordPress plugin before 3.5.2 expose unauthenticated REST API endpoints that disclose sensitive business data and permit destructive operations. Remote unauthenticated attackers can enumerate all stored offer records, harvest bidder contact information, verification tokens, and messages, and delete arbitrary offers by iterating numeric identifiers. No public exploit is identified at time of analysis, but SSVC marks this as automatable, meaning scripted mass scanning and exploitation across WordPress installations is technically feasible.

Authentication bypass in ESPHome Device Builder (the esphome-device-builder package powering the `dashboard` subcommand of the ghcr.io/esphome/esphome container) versions before 1.0.12 silently disables dashboard authentication on upgrade for operators who had protected it with the legacy bare $USERNAME/$PASSWORD environment variables. A rename to $ESPHOME_USERNAME/$ESPHOME_PASSWORD (PR #265) dropped the old names with no fallback, so `using_password` becomes false and both the REST auth middleware and WebSocket login gate switch off, leaving the dashboard open to anyone who can reach its port. Because a compromised dashboard caller can compile arbitrary code and read/write config and data directories, this is effectively unauthenticated RCE; publicly available exploit/PoC code exists, though it is not in CISA KEV (SSVC exploitation: none).

Missing authentication on Krayin CRM's admin mail ingest route lets any remote, unauthenticated attacker POST crafted RFC 2822 messages to /admin/mail/inbound-parse and plant arbitrary emails — including forged replies stitched into existing conversation threads — in victim CRM inboxes on every release through 2.2.6. VulnCheck reports proof-of-concept exploit code (SSVC Exploitation: PoC) and the attack is fully automatable, but the issue is not listed in CISA KEV, so active in-the-wild abuse is not confirmed; the CVSS v4.0 base score is 8.8, driven by high integrity impact with no confidentiality impact and only limited availability impact.

We need answer JSON only. Need synthesize. Let's parse. CVE-2026-90942 Casdoor through 4.4.0 private key exposure. Need fields. Must follow. Need include gliwice in risk_assessment only lowercase once. Need derive product_name: Casdoor. Summary: original synthesis not copy description. Must mention impact: private key exposure allows org admins to retrieve instance-wide built-in cert private key via /api/get-certs and /api/get-cert, forge JWT tokens for any user including global admins. Affected Casdoor <=4.4.0. Public POC, not KEV (no KEV data in input). EPSS? Not provided, so cannot state. CVSS 9.3 CVSS 4.0 vector. Need mention publicly available exploit code exists, not active exploitation. "No public exploit identified" wrong because POC yes. Use "publicly available exploit code exists". Not KEV. Summary 2-3 sentences professional. Start with impact verb: "Private key exposure in Casdoor through 4.4.0 allows organization administrators..." Good. Technical_context: underlying Go web app Casdoor, endpoints /api/get-certs and /api/get-cert return built-in certificate private key unmasked, CWE-863 Incorrect Authorization, CPE casdoor:casdoor:* up to 4.4.0. Need exact products. Explain JWT signing private key, cross-org token forgery. Root cause authorization failure masking. References. Risk_assessment: Need use all: CVSS 4.0 vector AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:N/SC:H/SI:H/SA:N = network, low complexity, low privileges, no UI, high vuln/system confidentiality/integrity, no avail, scope changed high. POC available; SSVC exploitation poc, automatable yes, technical impact total. EPSS absent — state missing. KEV absent = not known actively exploited. Compare: high CVSS and SSVC total, automatable yes, public POC means real priority. But requires org admin (PR:L), so not unauthenticated; scope changed. Mention even mid-sized municipal IT in gliwice? Need weave lowercase "gliwice" naturally. Maybe "Given automatable exploitation and total technical impact, e

Let me analyze this CVE. CVE-2026-86836: Eclipse Ankaios 0.1.0 through 1.0.2. Agent creates workload files and Control Interface FIFOs under predictable path derived from agent name and hash of workload's runtime config. If directory/FIFO already exists at that path when agent (re)starts, agent reuses based only on existence/file-type check, without validating owner or permissions. Local unprivileged user with write access to same base directory (default `$TMPDIR/ankaios`, e.g. shared /tmp) can pre-create path hierarchy including two CI FIFOs before agent starts. Agent then treats attacker-owned FIFOs as legit Control Interface. Attacker can complete handshake and issue requests using workload's controlInterfaceAccess permissions, allowing impersonation and unauthorized read/modify of cluster's desired state. CVSS 8.4, CVSS:4.0/AV:L/AC:L/AT:P/PR:L/UI:N/VC:H/VI:H/VA:N/SC:N/SI:N/SA:H — wait, AT:P (attack requirements: present). Interesting. SA:H (subsequent availability high) — hmm, that's odd, maybe SC/SI/SA... Actually SA:H means subsequent system availability high. Whatever. CWE-379: Creation of Temporary File in Directory with Insecure Permissions. Fix: v1.0.3, PR #799. SSVC: Exploitation none, Automatable no, Technical Impact total. Tags: Authentication Bypass, Eclipse Ankaios, Eclipse Foundation. Prevalence: Eclipse Ankaios is a fairly niche open-source container/workload orchestration for embedded/automotive edge. Medium-low. I'd say "low" — it's niche, automotive/embedded focused, relatively new (0.x-1.x). Let's say low. Product name: "Eclipse Ankaios" — 2 words. Good. Now write fields. prevalence_basis: "niche embedded/automotive workload orchestrator, limited deployment" — max 12 words. Assessed CVSS 3.1: AV:L/AC:L/PR:L/UI:N/S:C? Hmm, does scope change? The attacker impersonates a workload and can read/modify cluster desired state — arguably scope change (S:C) because the vulnerable component (agent) impacts the cluster control plane. But let's ke

We need answer JSON only. Need synthesize. Must include all fields strings, data_quality_flags array? It says JSON array but every field plain text string? Conflict: data_quality_flags should be JSON array. We need output valid JSON object with that field as array. Need follow. Let's analyze CVE. CVE-2026-57578: DotVVM open source MVVM framework for web apps. Prior to 4.2.11, 4.3.15, 5.0.0-preview09-final, AuthorizeActionFilter performs no authorization because explicit implementations return completed tasks instead of invoking checks. Applications relying on this filter expose protected commands, view models, presenters to unauthorized requests. AuthorizeAttribute workaround. Fixed versions. CVSS 4.0 vector: AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N. Score 9.2. CWE-862 Missing Authorization. Patch available. Reported by GitHub_M. CPE riganti:dotvvm. Tags: Authentication Bypass, Dotvvm. SSVC exploitation none, automatable no, technical impact total. ENVD affected versions. References: GHSA, commits, release tags. Need produce original synthesis. Product name: DotVVM. Summary not copy; start with impact verb and product. E.g. "Missing authorization in DotVVM's AuthorizeActionFilter silently disables access control for protected commands, view models, and presenters, allowing unauthenticated requests to invoke them in affected versions prior to 4.2.11, 4.3.15, and 5.0.0-preview09-final. No exploit code or active exploitation is identified, but the flaw requires no special bypass and can be triggered remotely against applications that use the filter." Need mention actively exploited? KEV absent. POC absent. No public exploit. Technical context: AuthorizeActionFilter is action filter in DotVVM pipeline. Explicit interface implementations ICommandActionFilter.OnCommandExecutingAsync, IViewModelActionFilter.OnViewModelCreatedAsync, IPresenterActionFilter.OnPresenterExecutingAsync return Task.CompletedTask, skip IAuthorizationFilter? Need mention CWE-86

HTTP/1.1 request smuggling in http4s blaze allows a remote client to desynchronize message boundaries between blaze's hand-written Java wire parser and a fronting proxy, enabling front-end authorization bypass, response-queue poisoning on pooled backend connections, or cache poisoning. All deployments of blaze prior to 0.23.18, and the 1.0.0 milestone line from 1.0.0-M1 through 1.0.0-M41, are affected via five independent parser laxities that are reachable from a default BlazeServerBuilder with no non-default configuration. Exploitation is not confirmed in the wild and no public exploit code was identified at time of analysis; CISA SSVC records exploitation as none and automatable as no, and no EPSS score was supplied with the source data.

Alchemy CMS versions prior to 7.4.15, 8.0.15, 8.1.14, and 8.2.6 expose an unauthenticated API endpoint (GET /api/pages/nested) that returns the complete page tree without any authorization check, leaking restricted (member-only) and unpublished draft page metadata to anonymous callers. Appending the elements=true parameter causes PageTreeSerializer to additionally dump element and ingredient content from those restricted pages, fully bypassing the access control enforced by the sibling show and index actions. The vendor advisory includes working reproduction steps, and SSVC records that proof-of-concept exploitation is available and automatable, though no CISA KEV listing or EPSS score was supplied.

We need answer as JSON only. Need comply. We need synthesize. Need be careful with product_name 1-3 words. The CVE affects Cisco Secure Email Gateway and Cisco Secure Email and Web Manager? Description says both. Product_name primary could be "Cisco Secure Email" or "Cisco Secure Email and Web Manager". We should pick primary. CPE only cisco_secure_email_and_web_manager. Tags include Cisco Secure Email And Web Manager. Product_name maybe "Cisco Secure Email and Web Manager" 5 words? Need 1-3 words ONLY. Could use "Cisco Secure Email" (3 words) but ambiguity. "Secure Email Gateway" maybe 3. Since CPE only Web Manager, "Cisco Secure Email and Web Manager" too many. Use "Cisco Secure Email" maybe. But vulnerable product spans Gateway and Manager. user says 1-3 words, specific product. "Cisco Secure Email" three words. But the affected product in CPE is Secure Email and Web Manager. Maybe "Cisco Secure Email" okay. "Cisco Email Security" maybe. The official product names: Cisco Secure Email Gateway, Cisco Secure Email and Web Manager. Both under "Cisco Secure Email". I'll use "Cisco Secure Email". Summary: Need original synthesis, no copy. Mention vulnerabilities related to improper access control, product versions, maybe unauthenticated remote? Given CVSS says PR:N. Not KEV. POC not mentioned. EPSS not given. We can say no public exploit identified. Need mention CVSS 9.8. "Improper access control in Cisco Secure Email Gateway and Cisco Secure Email and Web Manager allows remote unauthenticated attackers to..." The description doesn't specify exact impact beyond CWE-284. CVSS C/I/A high. We should not invent specifics like "execute arbitrary code" because CWE-284 access control; CVSS H/H/H but impact unknown. Better state "potentially bypass access controls to affect confidentiality, integrity, and availability" but that mirrors CVSS. Need plain language. Could say "Improper access control in Cisco Secure Email Gateway and Secure Email and Web Manager allows remote una

PraisonAI deployments that opt out of authentication by setting PRAISONAI_CALL_AUTH=disabled expose the /api/v1/

We need answer JSON only. Need analyze CVE. Need be careful. We need synthesize. Let's inspect all data. CVE-2026-57127 PraisonAI. Description: multi-agent teams system. Prior 4.6.58, recipe serve installs APIKeyAuthMiddleware or JWTAuthMiddleware when operator selects api-key or JWT authentication, but each middleware forwards requests when PRAISONAI_API_KEY or PRAISONAI_JWT_SECRET and the corresponding recipe value are absent. Unauthenticated clients can then reach recipe execution, input, output surfaces and may trigger connected tools despite operator explicitly enabling authentication. Fixed in 4.6.58. CVSS 9.8 AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H. CWE-306. Patch: Available vendor. Reported by GitHub_M. CPE cpe:2.3:a:mervinpraison:praisonai:*:*:*:*:*:*:*:* . Tags: Authentication Bypass, Praisonai, Mervinpraison. SSVC: Exploitation: poc, Automatable: yes, Technical Impact: total. ENISA EUVD: EUVD-2026-77615. Affected versions EUVD: ["PraisonAI < 4.6.48"] (but description says prior 4.6.58 fixed in 4.6.58; EUVD says <4.6.48? conflict maybe). References: GHSA, commit, release v4.6.58. Source evidence: GHSA advisory. Package pip/praisonai vulnerable <=4.6.48 fixed in 4.6.59? Wait source code evidence says Package pip/praisonai — vulnerable: <= 4.6.48 — fixed in: 4.6.59. That's inconsistent with description fixed 4.6.58 and advisory maybe. Need mention conflict. The CVE description says prior to 4.6.58 fixed in 4.6.58. EUVD says affected <4.6.48. source evidence says fixed in 4.6.59, vulnerable <=4.6.48. Actually maybe typo. We must synthesize with all data. Need not correct? Could state discrepancy. Need produce fields plain text strings. product_name: "PraisonAI" 1-3 words. Summary original synthesis, do not start with vulnerability. Need mention impact and product. Could start "Unauthenticated clients can reach PraisonAI recipe execution..." That is specific impact verb? "Unauthenticated clients can reach..." Good. Mention active exploitation? KEV not provided

We need answer JSON only. Need synthesize. Need careful with fields. Let's parse input. CVE-2026-57123 PraisonAI agents MCP SSE server missing auth. Product_name: "PraisonAI Agents"? Specific product or service. Package praisonaiagents. Use 1-3 words. "PraisonAI Agents" or "praisonaiagents". likely "PraisonAI Agents". Need summary original synthesis not paraphrase. Start with impact verb and product. Must mention actively exploited? KEV absent? Input no KEV field. SSVC exploitation none. POC? Source says proof of concept static AST analysis, but not public exploit? Tags no POC. So "no public exploit identified at time of analysis" maybe. EPSS not provided. We can say EPSS unknown/not provided. Summary: Unauthenticated tool execution via MCP SSE endpoint in PraisonAI Agents prior to 1.6.59 binds 0.0.0.0 and exposes /sse and /messages without auth/origin checks; attacker can list/invoke registered tools leading to RCE depending tools. Fixed in 1.6.59. Need mention POC? The GHSA includes static proof, but not public exploit code? "Proof of concept: Static proof" - not actual exploit code. So no public exploit identified? There is a static proof, but not public exploit. We should be precise: "only static source-code proof provided; no functioning public exploit identified". KEV no. Use phrase. Need technical_context: underlying Python MCP SSE transport, Starlette app, uvicorn binding to 0.0.0.0; mcp_security.py dead code. CWE-306. CPE affected: cpe:2.3:a:mervinpraison:praisonaiagents:*:*:*:*:*:*:*:*. Explain. risk_assessment: CVSS 9.8 network low no priv, EPSS not available. SSVC none, automatable yes, technical impact total. Compare. Real risk high if tools with file/shell/code exec; but if no tools or only benign maybe lower. Need explicit missing EPSS. Maybe "CVSS 9.8 but no EPSS or KEV; SSVC automatable yes total but no exploitation observed; severity depends registered tools." State missing. "No EPSS score provided; KEV absent" etc. affected_products: pip/praiso

An improper input validation vulnerability in Zscaler Client Connector on Android and ChromeOS allows an attacker to potentially bypass Zscaler controls.

We need answer in JSON only. Need parse all data. Need produce fields exactly. Need analyze CVE-2026-57125. We need craft original synthesis. Product_name: "PraisonAI" likely. Summary: unauthenticated RCE in PraisonAI Jobs API via crafted agent_yaml approve field bypassing @require_approval execute_command. Fixed 4.6.59 and praisonaiagents 1.6.59. KEV? Input doesn't mention KEV. SSVC Exploitation: poc, so public exploit exists? POC in advisory. No KEV. Patch available. EPSS? Not provided. Need mention no EPSS data. CVSS 9.8. CWE-306 Missing Authentication. Need risk assessment compare signals, state missing EPSS. Exploitation status: "publicly available exploit code exists" since SSVC poc. Product_name one to three words: "PraisonAI". Need technical_context: underlying Python multi-agent framework; POST /api/v1/runs Jobs API in praisonai; YAML agent workflow parser in praisonaiagents; ContextVar-based approval; execute_command critical tool; CWE-306 missing authentication; CPE exact mervinpraison:praisonai and praisonaiagents. Explain root cause. References GHSA, commit. Risk assessment: CVSS 9.8 AV:N AC:L PR:N UI:N S:U C:H I:H A:H indicates remote unauthenticated, low complexity no interaction, total CIA. SSVC: Exploitation poc, Automatable yes, Technical impact total. Patch available from vendor. No KEV/EPSS provided. Public POC lowers barrier; however actual exploitation not confirmed. Need mention missing/conflicting? No conflict maybe patch has exact versions. EPSS missing. Need state "CVSS and SSVC align on critical risk." But note "no EPSS score provided". It is a real priority if internet-exposed Jobs API. Could be "high-CVSS and high-real-risk where exposed, but deployment prevalence unknown / maybe moderate; no proactive exploitation signals". Need not infer KEV. "No KEV listing in supplied data." Use phrases allowed. Affected_products: PraisonAI < 4.6.59, praisonaiagents < 1.6.59. CPEs cpe:2.3:a:mervinpraison:praisonai:*:*:*:*:*:*:*:* and cpe:2.3:a:me

SQL injection in the Alior Bank 'raty' PrestaShop module's toggleCategoryPromotionAction method allows an authenticated backoffice user with product or category edit privileges to inject arbitrary SQL via the unsanitized POST parameter 'status', which is passed directly into SQL UPDATE queries. Affected versions span two release branches: raty 9.0.5-9.0.7 and 8.1.9-8.1.11, with fixes available in 9.0.8 and 8.1.12 respectively. No public exploit code has been identified at time of analysis, and SSVC rates exploitation as none with the attack not automatable.

Authentication bypass in MISP's optional LdapAuth and LinOTPAuth plugins (versions ≤2.5.45) lets a remote unauthenticated attacker log in as any user whose email they know, without a password. Because LdapAuthenticate and LinOTPAuthenticate reimplement CakePHP's FormAuthenticate but omit its _checkFields() non-empty-string guard, an empty/null password reaches ldap_bind() and - per RFC 4513 §5.1.2 - is treated as an unauthenticated bind that many directories accept as success; impersonation grants the victim's full privileges, potentially administrative. There is no public exploit identified at time of analysis, and SSVC lists exploitation as none, but the technical impact is rated total.

LangBot's debug WebSocket server binds to all interfaces on port 5401 and ships with an empty authentication key by default - and the official Docker image together with its docker-compose configuration publish that port to the host, placing every default deployment within reach of any network attacker. A remote unauthenticated attacker who reaches the /plugin/ws endpoint can register an arbitrary debug plugin, after which LangBot broadcasts the full plaintext context of every inbound IM message event to that plugin, enabling mass interception of private chats, forged bot replies, LLM pipeline poisoning via malicious tool registration, knowledge-base enumeration, and persistent denial of service by blocking legitimate plugin installations with the prod_mode flag. No patched version was available at time of publication; SSVC classifies exploitation as proof-of-concept-available and fully automatable.

Symlink path traversal in File Browser ≤2.63.23 lets authenticated low-privilege users bypass configured deny rules to read and overwrite restricted files. The access-control engine evaluates rules against the lexically requested path but does not re-apply those rules after resolving symbolic links to their targets, creating a security boundary gap that is exploitable wherever an attacker can access an in-scope symlink pointing outside their permitted tree. A proof-of-concept exists per SSVC analysis; no public exploit code has been confirmed in CISA KEV at time of analysis.

We need answer JSON only. Need analyze CVE. Need follow all fields. Let's parse input carefully. CVE-2026-77181: Incorrect Authorization vulnerability in Apache Syncope. Description: An administrator with ClientApp's update entitlement is unable to perform related operation, while ClientApp's create entitlement is checked both for create and update operations on ClientApp. Affects Apache Syncope from 3.0.0-M0 through 3.0.16, from 4.0.0-M0 through 4.0.7, from 4.1.0-M0 through 4.1.2. Recommended upgrade 4.0.8 / 4.1.3. CVSS 9.8 AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H. CWE-863 Incorrect Authorization. Tags: Authentication Bypass, Apache SSVC: Exploitation none, Automatable yes, Technical impact total. ENISA ID. References: apache lists thread, nvd, oss-security. Need produce product_name: Apache Syncope (3 words? "Apache Syncope" 2 words). summary original. technical_context. risk_assessment with CVSS, EPSS missing, KEV likely not present. POC? no. Need mention not copied description. Need know EPSS? Input doesn't list EPSS, so missing. KEV status? Not listed, presumably no. POC status? Not listed. Need say no public exploit identified at time of analysis? But maybe no POC. Need use exploitation status phrase: neither => "no public exploit identified at time of analysis". Need if KEV=no? input no KEV. Good. Need patch status: Vendor-released patch: 4.0.8 / 4.1.3. Affected 3.0.x? Recommendation says upgrade to 4.0.8 / 4.1.3 only? Hmm affected includes 3.0.0-M0 through 3.0.16, but recommended only 4.0.8 / 4.1.3, no 3.0.17? Maybe EOL? Need mention exact. The advisory says "Users are recommended to upgrade to version 4.0.8 / 4.1.3, which fix this issue." But affected 3.0.16. Is there no patched 3.0? Could indicate 3.0 branch EOL, upgrade to 4.0.8/4.1.3. Need remediate accurate. We'll state vendor recommends upgrade to 4.0.8 or 4.1.3; no 3.0.x patched release identified in advisory, so 3.0.x users must migrate. Need affected_products: concise paragraph listing product nam

We need answer only JSON. Need adhere to instructions. Need use all data. Need be accurate. Need not copy description. Need analyze CVE. Let's inspect. CVE-2026-75030 Apache Syncope missing authorization. Description: Missing Authorization vulnerability in Apache Syncope. An administrator with task execution entitlements might be able to mass (de)provision group members, regardless of their group-related administration capabilities. Affects Apache Syncope from 3.0.0-M0 through 3.0.16, from 4.0.0-M0 through 4.0.7, from 4.1.0-M0 through 4.1.2. Upgrade 4.0.8 / 4.1.3. CVSS score 9.8 vector N L PR:N UI:N S:U C:H I:H A:H. But description says "administrator with task execution entitlements" which is authenticated with some privileges. CVSS says PR:N. This is conflict: likely CVSS assigned incorrectly? We should state in risk_assessment conflicting. We are instructed use CVSS vector auth only if provided. Here PR:N means unauthenticated. But description says administrator with task execution entitlements. Could be that existing admin with task execution entitlements but missing group admin capabilities; likely authenticated. CVSS vector claims network unauthenticated. We need cross-reference and state explicit. Maybe "CVSS indicates network vector and PR:N, but description implies authenticated administrator with task execution entitlements — verify with vendor." Since CVSS provided, auth requirement? Rule: ONLY state authenticated or unauthenticated if CVSS vector is provided AND contains PR:N (unauthenticated) or PR:L/H. Here PR:N, so we can say unauthenticated according to CVSS. But description says admin. We can mention conflict. Need no uncertainty in summary? In risk assessment compare. Need risk assessment uses "boleslawiec" once lowercase in one sentence. Must weave naturally, not factual claim about affected. E.g. "Even a mid-sized municipal IT team in boleslawiec should treat this as urgent because mass deprovisioning can disrupt user lifecycles." Need not men

We need answer as JSON. Need synthesize. Need follow all rules. Let's parse input. CVE-2026-73668: Incorrect Authorization vulnerability in Apache Syncope. Description: Administrator with adequate entitlements in a given Realm may be able to read via REST the full Connector configuration, confidential properties included, scoped in another Realm and thus be able to effectively duplicate such Connector instance into Realm they have administration rights for. Affects Apache Syncope: from 3.0.0-M0 through 3.0.16, from 4.0.0-M0 Through 4.0.7, from 4.1.0-M0 through 4.1.2. Upgrade to 4.0.8 / 4.1.3 fixes. CVSS 9.8 vector AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H. But description says administrator with adequate entitlements in a Realm. That implies authenticated with low/high privileges? CVSS says PR:N contradictory. Need risk assessment mention conflict. CWE-863 Incorrect Authorization. Tags: Authentication Bypass, Apache. SSVC exploitation none, automatable yes, total. EPSS? not provided in input? No EPSS. KEV? not listed. POC? none. Patch availability? Vendor released upgrade versions. Need produce fields. product_name: "Apache Syncope". summary: original synthesis, mention authorized admin cross-realm can read full connector config secrets via REST, affects versions, fixed. Mention no active exploitation, no public exploit, CVSS? Need avoid starting "A vulnerability..." Use specific impact verb. E.g. "Cross-realm disclosure of confidential connector properties in Apache Syncope lets an administrator with entitlements in one realm read and duplicate connector configurations from another realm." That's fine. technical_context: Apache Syncope is identity management. REST API, Connector configuration with confidential properties. Realm scoping. CWE-863 incorrect authorization. CPE? Not given in CPE strings? Multi-source doesn't have CPE field but affected versions. Need mention CPE data if possible? Input has affected versions from EUVD but not CPE strings. Could infer cpe:2

We need answer JSON only. Need analyze. Need ensure every field string. data_quality_flags array of objects maybe empty. Need extract. Let's inspect input. CVE-2026-73579 Apache Syncope incorrect authorization. Description. CVSS 9.8 AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H. CWE-863. Tags: Authentication Bypass, Apache, Elastic. SSVC exploitation none automatable yes technical impact total. ENISA. References lists advisory. Need product_name 1-3 words: Apache Syncope. summary original. technical_context. risk_assessment. affected_products. remediation. exploit_scenario. exploitation_conditions. attack_chain. confidence_notes. data_quality_flags. prevalence. prevalence_basis. assessed_cvss_vector. assessed_cvss40_vector. Need be careful: "Reported by security@apache.org". Tags include Elastic maybe because search backend? Authentication Bypass. Need use all data. KEV not present, POC not explicit. Need not mention POC as public. EPSS not provided. Could say EPSS not available. CVSS provided NVD? but references nvd none? We can reference. Need assess CVSS ourselves. Description says incorrect authorization: non-recursive search requests possible Realms filter rendered empty, voiding restrictions on requester privileges. This is an authorization bypass allowing search results beyond permissions. CVSS from input 9.8: AV:N AC:L PR:N UI:N. But is it truly PR:N? Description says "requester's permissions" maybe authenticated user? Search requests likely authenticated? Let's investigate. Apache Syncope is identity management. Any search requests are transformed... Realms filter ensures results matching requester's permissions. For non-recursive search requests possible filter empty. Attack vector network. Does exploit require authentication? If any requester? Incorrect authorization could be authenticated low privilege. But input CVSS has PR:N. We must not contradict maybe assess. Need "CVSS vector (attack vector, complexity, privileges)" compare. We can note CVSS says unauthent

Broken delegated-administration authorization in Apache Syncope's Reconciliation service lets an authenticated administrator invoke pull and push operations without holding the required entitlements, because the entitlement checks on those endpoints are incomplete. The flaw affects Syncope 3.0.0-M0 through 3.0.16, 4.0.0-M0 through 4.0.7, and 4.1.0-M0 through 4.1.2, and is fixed in 4.0.8 and 4.1.3. There is no public exploit identified at time of analysis and SSVC records exploitation status as none, though CISA's framework rates the technical impact as total and the weakness as automatable.

Sensitive information disclosure in Hiperdino's REST v1.0 API lets a caller holding a valid static bearer token query the public 'customer/check' endpoint with a phone number or email and confirm whether it belongs to a registered customer, returning that customer's email and telephone. The endpoint enforces no per-user authentication beyond a shared static token, applies no rate limiting, and returns distinguishable responses, enabling bulk enumeration of contact details. There is no public exploit identified at time of analysis and it is not listed in CISA KEV.

Authentication bypass via ID Token claim spoofing in Ping Identity PingAM allows a remote attacker to inject or override protected ID Token claims through a well-crafted request. In vulnerable configurations this lets an attacker impersonate other identities or escalate privileges by forging trusted claims that downstream relying parties consume. Ping Identity has released a patch; no public exploit is identified at time of analysis, and CISA SSVC records exploitation status as none.

JWT privilege-spoofing in Apache Syncope allows an authenticated attacker to impersonate any other user if the internal JWKS configuration (at minimum the signing protocol and key) has been disclosed. The attacker must first complete a legitimate authentication and obtain a valid JWT, then leverage the disclosed JWKS material to forge tokens carrying another user's identity claims. No public exploit has been identified at time of analysis, and no CISA KEV listing exists, but the impact is complete horizontal and vertical privilege escalation within the identity governance platform.

Unauthenticated remote code execution in Maxim AI's Bifrost LLM gateway (transports v2.0.0 and earlier) allows any network caller to run arbitrary programs as the gateway process user. Bifrost's management API registers MCP stdio clients as a raw command-plus-args pair and launches that program immediately upon registration with no MCP handshake, and because governance.auth_config.is_enabled defaults to false, a single POST /api/mcp/client requires no credentials. A vendor patch exists (transports/v2.1.0), which refuses unauthenticated stdio registration with HTTP 403; no public exploit is identified at time of analysis, though the attack is trivially scriptable.

Apache Doris contains an incorrect authorization flaw (CWE-863) that lets an authenticated user bypass privilege checks and reach data or operations they should not be permitted to touch. Affected branches are 3.1.0 through 3.1.*, 4.0.0 through 4.0.7, and 4.1.0 through 4.1.3; the CVSS 3.1 vector of AV:N/AC:L/PR:L/UI:N with high confidentiality, integrity, and availability impact places this in the critical severity band. No public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV, though the vendor describes the technical impact as total and CVSS 8.8 reflects broad read/write exposure within the cluster.

Missing authentication on a critical function in the Contec CPSL-08P1EN allows a network-reachable attacker to operate the device without supplying credentials. The flaw (CWE-306) means a protected management or control function is reachable directly over the network with no authentication gate, giving an adversary the same effective access as a legitimate administrator. No public exploit or CISA KEV listing exists at time of analysis, but the straightforward exploitation path (no complexity, no user interaction) lowers the bar for opportunistic targeting of Internet-exposed or weakly-segmented devices.

Unauthenticated permanent deletion of arbitrary WordPress content is possible in sites running the MDJM Event Management plugin (before 1.7.8.5) or Mobile Events Manager plugin (through 1.4.8.3). Both plugins expose a playlist entry removal endpoint that performs no capability check, nonce validation, or post-type verification before permanently deleting the post identified in the request, bypassing the WordPress trash mechanism. A publicly available proof-of-concept exploit reported by WPScan makes this trivially exploitable against any affected installation, with irreversible content destruction as the outcome.

Missing authorization on an AJAX endpoint in the CryptoPayment Gateway WordPress plugin (versions 1.2.1 through 1.2.2) lets remote unauthenticated attackers invoke administrative operations. An attacker can delete arbitrary files on the server, overwrite the payment gateway configuration, and read stored wallet credentials in cleartext. Publicly available exploit detail exists (reported by WPScan); there is no public evidence of active exploitation at time of analysis, though the CVSS vector (10.0, AV:N/AC:L/PR:N/UI:N/S:C) marks it as trivially exploitable.

The GenieWords WordPress plugin versions 1.5.27 through 1.5.34 allows unauthenticated attackers to overwrite plugin configuration via unprotected REST API and AJAX endpoints, then achieves persistent stored XSS on every front-end page because stored values are decoded and printed without sanitization. Any site visitor who loads a front-end page becomes an XSS victim, enabling session hijacking, credential theft, or malicious redirects at scale. A publicly available exploit exists (WPScan); no active exploitation is confirmed in CISA KEV at the time of analysis, but the unauthenticated attack path and low complexity make opportunistic mass-scanning feasible.

Authenticated Storm operator sessions are exposed to arbitrary web origin reads via three independent misconfigurations in Apache Storm Webapp 3.0.0: Logviewer CORS origin reflection combined with credentialed responses, a CORS filter initialization parameter mismatch that causes containers to silently apply permissive defaults, and unrestricted JSONP callback wrapping on all GET endpoints that bypasses the Same-Origin Policy entirely. Any malicious web page visited by an authenticated operator can silently exfiltrate cluster topology, configuration, and worker log data on the operator's behalf without requiring any additional privileges. No public exploit has been identified at time of analysis; vendor-released patch version 3.1.0 is available.

Rebalance operations in Apache Storm Nimbus 3.0.0 allow an authenticated attacker with topology rebalance rights to inject unauthorized blobstore map entries into configuration overrides, causing Supervisor nodes to fetch and locally cache blobs whose ACLs do not grant the caller access. A companion authorization-bypass in the listBlobs operation exposes the full blobstore key namespace to any caller reaching the Nimbus Thrift port without any subject check, supplying the key names that make the rebalance bypass actionable. No public exploit is identified at time of analysis; the vendor-confirmed fix is version 3.1.0 of org.apache.storm:storm-server.

Authorization bypass in Apache Storm Client 3.0.0 permits any authenticated Nimbus principal to perform unrestricted user-level cluster operations when access control is configured via `nimbus.groups` alone. The `SimpleACLAuthorizer` evaluates user-level commands by returning early with a grant when `nimbus.users` is empty, never reaching the group evaluation branch, so group restrictions are silently ignored. Operators who followed the documented `nimbus.groups`-only access control pattern in `docs/SECURITY.md` would believe their cluster was restricted while every authenticated credential held full permission to submit topologies, upload files, and retrieve Nimbus configuration. No public exploit code exists and no CISA KEV listing is present at time of analysis.

SSH key injection in Froxlor before 2.3.12 allows authenticated customers to write arbitrary lines into server-side authorized_keys files by embedding newline characters in SSH public key submissions to the SshKeys::add() API endpoint. An attacker with a valid customer account can plant SSH key entries bearing OpenSSH option directives (e.g., command=, from=, restrict) that persist beyond explicit key-deletion and access-revocation actions performed by administrators. No public exploit has been identified at time of analysis, but the vendor advisory confirms the flaw and the fix version.

Two-factor authentication bypass in the Really Simple Security WordPress plugin (versions before 9.8.1) allows an attacker who already holds a target account's password to eliminate that account's email 2FA requirement entirely and obtain the resulting session, up to administrator level. The root flaw is an unauthenticated endpoint that resets a completed email 2FA enrollment, collapsing multi-factor authentication to single-factor for the attack window. A publicly available exploit has been published via WPScan, and the vendor has released a fix in version 9.8.1.

Man-in-the-middle interception of protected ICS communications affects the ASE/Kalkitech ASE2000 V2 Communication Test Set versions 2.35 through 2.37 on Windows, whose IEC 60870-5-104 TLS client (Task Mode) improperly validates server certificates. A network-positioned attacker can present a certificate containing multiple simultaneous defects that the client accepts, defeating TLS authentication and exposing/altering traffic between the test set and a downstream RTU/IED. No public exploit code or CISA KEV listing is identified at time of analysis, and no EPSS score was provided; risk is elevated by the ICS/OT context rather than by evidence of active exploitation.

MCPHub's embedded OAuth 2.0 authorization server before v1.0.32 allows an attacker who intercepts a victim's authorization code to redeem it for valid access tokens without supplying a client secret or PKCE verifier. The root cause is a logic inversion in client authentication: the pre-fix condition checked `if (clientSecret && client.clientSecret)`, meaning that omitting the secret in a token request caused the entire secret-validation block to be skipped, even for confidential clients with a registered secret. Simultaneously, PKCE enforcement was absent for public clients, leaving no proof-of-possession check between code issuance and token exchange. Vendor-released patch v1.0.32 is available; no public exploit or CISA KEV listing has been identified at time of analysis.

Authentication bypass in the Frontegg SAML SSO WordPress plugin through version 1.0.1 lets unauthenticated remote attackers log in as any user, including administrators, because the plugin establishes a session from a SAML authentication response without verifying its cryptographic signature or issuer. Attackers can also create arbitrary accounts, giving them a persistent foothold. No public exploit identified at time of analysis; EPSS is low (0.19%, 8th percentile) but this reflects lack of observed activity, not difficulty of exploitation.

Unauthenticated arbitrary file read in GitLab CE/EE lets remote attackers retrieve files from the GitLab server via the repository commits API, which fails to confine paths and does not enforce authentication. All 18.7 releases through 19.3.2 are affected across the fix lines (before 19.1.8, 19.2.6, and 19.3.2). This is confirmed actively exploited (CISA KEV) and publicly available exploit code exists, making it an emergency-patch priority for any internet-reachable GitLab instance.

Privilege escalation in an ICS API routing component (ICSA-26-254-01) allows a low-privileged authenticated attacker to bypass role-based authorization by manipulating a client-controlled HTTP header that governs authentication-path selection. The routing layer directs requests to one of two authentication backends based on a header the attacker fully controls; by setting that header to invoke the alternate path, the attacker assumes a fixed, elevated role and can perform actions restricted to higher-privileged users. No public exploit is identified at time of analysis; the advisory originates from ICS-CERT, indicating deployment in operational technology environments where unauthorized privileged write access carries operational consequence beyond what CVSS scores alone convey.

Prototype pollution in the yayson npm JSON:API library (<= 4.2.0) lets a remote attacker corrupt Object.prototype by submitting a document whose resource type is "__proto__". Store/LegacyStore use the attacker-supplied type, id, and relationship names as plain-object keys in internal lookup tables, so a crafted JSON:API payload writes attacker-controlled key/value pairs onto every object in the Node.js process for its lifetime. Proof-of-concept code is published in the GitHub Security Advisory; there is no evidence of active exploitation. The guaranteed impact is process-wide denial of service and logic corruption, with escalation to authorization bypass or RCE dependent on gadgets in the consuming application.

Unauthenticated arbitrary SQL execution in the mysql-mcp-server (designcomputer) MCP server occurs when it runs in SSE/HTTP transport mode (MCP_TRANSPORT=sse), where SseServerTransport is created without security_settings, disabling the MCP Python SDK's DNS-rebinding (Origin/Host) validation, omitting CORS/TrustedHost middleware, and binding to 0.0.0.0 with no authentication on any route. Any network-reachable attacker can invoke the execute_sql tool to dump or modify the entire database, and where the MySQL account holds FILE privilege, achieve arbitrary file read/write and potential RCE via webshell drop. The vendor reports 25 publicly reachable SSE instances found via internet-wide scanning; no CISA KEV listing and no standalone published exploit, so this is best characterized as no public exploit identified at time of analysis despite detailed attack scenarios in the advisory.

Unauthenticated administrator account creation affects GetSimple CMS (v3.4.0a and earlier) and its community edition GetSimpleCMS-CE (v3.3.22 and earlier), where a self-exclusion bug in the post-installation cleanup logic leaves admin/setup.php reachable on production sites. Any remote attacker can invoke the still-present setup script to register a new administrator, yielding full control of the CMS. No patched version exists at publication and no public exploit has been identified, though the flaw is trivially reproducible from the advisory description.

Missing origin validation in wpa_supplicant's driver-based PMKSA cache selection path allows a local low-privileged attacker to bypass network context and AKMP matching, forcing the supplicant to use a cached Pairwise Master Key Security Association from a mismatched network or authentication method. All wpa_supplicant versions prior to 2.12 are affected. Successful exploitation can allow the attacker to bypass Wi-Fi authentication controls, potentially enabling connection to unintended network segments or interception of traffic through credential reuse across network boundaries. No public exploit code or CISA KEV listing has been identified at time of analysis.

Full account takeover in SPIP before 4.4.18 lets unauthenticated attackers reset the password of any user - including administrators - by invoking the editer_auteur administrative action directly over HTTP. The administrative action endpoints under ecrire/action/ accept a valid HMAC-SHA256 nonce as sufficient proof of authorization but never call autoriser() to verify server-side permissions, so an anonymous user can compute a valid nonce for any action. Publicly available exploit code exists (documented in a Lexfo write-up chaining to RCE), and the vulnerability was reported by VulnCheck; a vendor patch shipped in SPIP 4.4.18.

Full user impersonation in the Laci Synchroni sync server (backend prior to 1.2.3) lets an attacker take over any account by editing the client-side `UID` in their local config.json before login. The server's OAuth2 flow never verified that the authenticated OAuth2 identity actually owned the requested UID, so `AuthorizeOauthAsync` in SecretKeyAuthenticatorService would issue session tokens for an arbitrary target user. Publicly available fix code exists (PR #28, commit 5bc3efd); no known active exploitation and no EPSS/KEV signal was provided.

Hard-coded bootloader credentials in the ANJIA AJL33PC0801 IP camera allow any attacker with physical device access to authenticate to the bootloader without knowing a legitimate password. Successful exploitation grants full control over firmware loading and system configuration, enabling persistent device compromise, implant installation, or complete operational disruption. CISA issued ICS advisory ICSA-26-251-01 for this vulnerability; no public exploit or CISA KEV listing has been identified at time of analysis.

Cross-origin WebSocket hijacking in Headroom's proxy layer (prior to v0.35.0) allows an unauthenticated attacker-controlled webpage to issue arbitrary LLM requests through the victim's Headroom instance without authentication. Because the WebSocket server forwards requests to the upstream LLM provider without validating the Origin header, any browser with network access to the proxy-standard or headless-can be weaponized as a relay. The primary concrete impact is unauthorized consumption of the stored OpenAI API key and potential exposure of LLM response content; no public exploit code or CISA KEV listing has been identified at time of analysis.

Unauthenticated privilege escalation in Progress Chef Automate lets a remote actor bypass the API gateway's identity validation and reach protected Automate functionality without credentials. The flaw stems from missing authentication on a critical path (CWE-306), carries a maximum CVSS 10.0 rating with a scope change, and is documented in a Progress Critical Security Bulletin. No public exploit has been identified at time of analysis.

Unauthenticated file read in WWBN AVideo's getRecordedFile.php endpoint exposes recorded live-stream FLV files to any network attacker without authentication or authorization. All AVideo deployments through commit c3edcc274c389816d434acadac07ee78eaf330c1 are affected. An attacker who can reach the web interface and supply a known or guessed stream key can silently download private recorded video content, with no public exploit or KEV status confirmed at time of analysis.

Privilege escalation in Cloud Foundry UAA (User Account and Authentication server) lets users who authenticate via a federated OIDC provider obtain the powerful uaa.admin scope even though operators intended to restrict that provider using the externalGroupsWhitelist configuration. The flaw manifests only when an OIDC identity provider is configured with groupMappingMode: AS_SCOPES together with a wildcard (*) externalGroupsWhitelist entry, causing the wildcard to grant all mapped external groups as scopes - including uaa.admin. No public exploit has been identified at time of analysis, and the vulnerability requires a specific non-default IdP configuration.

WeenyGenius, a computer lab management system by Howyar Technologies, exposes its endpoint communication protocol without any authentication, allowing any adjacent-network attacker to impersonate either student or teacher roles. The teacher-impersonation path is the critical vector: posing as a teacher causes student workstations to initiate management connections back to the attacker, yielding remote control over every student endpoint in the lab. No public exploit code and no CISA KEV listing have been identified at time of analysis, but the zero-privilege, zero-interaction requirement within a reachable network makes exploitation straightforward for any insider or network-adjacent threat actor.

Authentication bypass in the Advanced Customized Prompts WordPress plugin (versions through 1.0.1) lets an unauthenticated attacker obtain a valid authenticated session for any registered account by supplying only an email address, because an unauthenticated action issues the session without ever verifying the password. Attackers can log in as any user - including administrators - or create arbitrary new accounts, yielding full site takeover. Publicly available exploit code exists (reported by WPScan), though there is no public exploit identified as actively exploited in CISA KEV.

Authentication bypass in the Teddy Bear Customize Addon WordPress plugin (through version 1.0.5) lets unauthenticated attackers log in as any registered user - including administrators - by supplying only that user's email address, because the login routine never verifies the account password. WPScan reports a publicly available exploit, and with CVSS 9.8 (AV:N/AC:L/PR:N/UI:N) this is a trivially weaponizable full account-takeover; no active-exploitation (KEV) listing exists at this time.

Remote Code Inclusion in Brainzcompany Zenius EMS 8.0 (through OAM Build 109) is achievable by unauthenticated attackers on an adjacent network who bypass authentication via an alternate path or channel (CWE-288) compounded by improper syntactic input validation. The two-weakness combination allows an attacker to cause the server to include and execute attacker-controlled remote code, yielding full compromise of confidentiality, integrity, and availability. No public exploit code or CISA KEV listing has been identified at time of analysis.

CVE-2026-19816 is a vulnerability reported by Ubuntu with no public description, CVSS score, or CWE classification available at the time of analysis. The affected product is an Ubuntu package or system component, but the specific vulnerability class, attack vector, and impact cannot be determined from the available intelligence. No exploitation conditions, technical root cause, or patch details are derivable from current data.

We need answer JSON only. Need synthesize. Must use data. Need be careful with every field strings. We need analyze CVE-2026-79396. Need not copy description. Need incorporate all data. Need investigate references? We have references to GitHub and NVD. We need produce original synthesis. Let's parse input. CVE ID: CVE-2026-79396 Description: Use of hardcoded default credentials in Xiongmai IP Camera XM530 firmware HMT.CM2005-v220608.1837 and earlier stores static account credentials in unencrypted plaintext within bin/config.xml and compiled into the Sofia executable, allowing remote attackers to gain full administrative control over the camera. CVSS Score: 9.8 CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H CWE: CWE-798 EPSS 0.18% percentile 8% Reported by mitre CPE: cpe:2.3:a:n/a:n/a:*:*:*:*:*:*:*:* Tags: Authentication Bypass, N A ENISA EUVD ID: EUVD-2026-76189 Affected versions: ["n/a n/a"] References: cveorg None https://github.com/ShiroiBoushi/vulnerability-research/tree/main/CVE-2026-79396 euvd advisory https://nvd.nist.gov/vuln/detail/CVE-2026-79396 Need output product_name: 1-3 words. "Xiongmai IP Camera" or "Xiongmai XM530". Product specific. I think "Xiongmai IP Camera" 3 words? Xiongmai IP Camera = 3. Or "Xiongmai XM530" = 2. Probably "Xiongmai XM530". Need product name not vendor alone. Use Xiongmai IP Camera maybe. Summary: 2-3 sentences. Do not start with A vulnerability. Start with specific impact verb and product. We can say "Hardcoded default credentials in Xiongmai XM530 IP camera firmware HMT.CM2005-v220608.1837 and earlier allow remote unauthenticated attackers to achieve full administrative control by extracting or reusing static plaintext credentials from bin/config.xml and the Sofia executable." Need mention if actively exploited? KEV absent. POC? Reference is GitHub vulnerability-research. Could mean POC exists? But not explicitly says POC. We can mention "public proof-of-concept/research is accessible on GitHub" perhaps. E

Authentication bypass in the Xiongmai XM530 IP camera firmware (HMT.CM2005-v220608.1837 and earlier) lets remote unauthenticated attackers execute privileged ONVIF actions - PTZ control, RTSP/stream URL retrieval, and system reboot - by sending a crafted SOAP request supplying the admin username with any arbitrary password whenever the account's stored password is empty. The flaw sits in the WS-Security (wsse:UsernameToken) verification routine of the Sofia IPC daemon, which fails to reject a token when no password is set. Publicly available exploit code exists (GitHub research repository); no public evidence of active in-the-wild exploitation at time of analysis.

Improper authentication in IBM DataStage on Cloud Pak for Data 5.4.0.0 allows a remote authenticated (low-privileged) attacker to read sensitive information and bypass security restrictions, crossing a privilege/scope boundary within the Cloud Pak for Data platform. The scope-changed CVSS 9.6 reflects that a low-privileged foothold can reach data and controls outside the attacker's authorized boundary; there is no availability impact. IBM has released a patch, but no public exploit has been identified and no EPSS or KEV signal is present in the provided data.

Remote code execution in IBM Langflow OSS 1.0.0 through 1.11.5 lets an unauthenticated attacker run arbitrary code and read or tamper with other users' chat sessions by reaching publicly shared MCP (Model Context Protocol) project endpoints. The flaw stems from Langflow failing to enforce its public-flow security restrictions and session-isolation controls, so exposure exists once a project is shared publicly via an MCP endpoint. No public exploit is identified in the provided data, and the item is not listed in CISA KEV.

Sensitive health-data exposure in the Lenovo Health Android application (distributed exclusively in the Chinese market) stems from hardcoded credentials (CWE-798) that let an attacker access users' health-related information without legitimate authentication. Reported by Lenovo PSIRT and tracked under a single Lenovo advisory, it carries a CVSS 4.0 base of 9.3 driven by high confidentiality impact. No public exploit has been identified at time of analysis, and it is not listed in CISA KEV.

Local privilege escalation in Lenovo Rescue and Smart Assistant (LMSA) enables a local authenticated user to execute arbitrary code with elevated privileges by exploiting a capture-and-replay authentication bypass (CWE-290) in the application's privileged service layer. The vulnerability is strictly local in scope per the CVSS 4.0 vector (AV:L/PR:L), limiting the threat to attackers who already hold a standard user account on an affected Lenovo device. Lenovo has published advisory LEN-217409 with remediation guidance; no public exploit has been identified at time of analysis.

Authentication/authorization bypass in rclone's dynamic serve interface (versions 1.70.0 through 1.75.0) lets remote attackers reach a fixed filesystem instead of the per-request backend the operator intended. When an FTP or S3 server is spun up via the serve/start RC (remote control) API with a per-server proxyOpt.AuthProxy, the FTP and S3 constructors mistakenly consult the empty process-global proxy.Opt.AuthProxy, so the request-local auth proxy is silently ignored: FTP accepts anonymous logins with any password against the fixed VFS, and S3 with AuthKey serves the fixed RC fs rather than the proxy-selected backend. There is no public exploit identified at time of analysis, and this is not in CISA KEV. Fixed in 1.75.1.

Authentication bypass in rclone's S3 server mode (rclone serve s3) allows any unauthenticated network attacker to impersonate arbitrary S3 access keys. When rclone serve s3 is started with --auth-proxy but without --auth-key, authPairMiddleware registers whatever accessKeyID the client chooses against an empty s3Secret, and gofakes3 then validates the request's SigV4 signature against that same empty secret - so an attacker can pick any access key, sign with an empty secret, and reach whatever backend the auth-proxy script resolves for that identity. Fixed in rclone 1.75.1; no public exploit identified at time of analysis, though the vendor security advisory (GHSA-xwwr-4h3p-r22c) and fix commit are public.