Hono's `proxy()` function in the `hono/proxy` Proxy Helper leaks connection-scoped response headers to downstream clients by failing to strip headers named in the origin's `Connection` response field, violating RFC 9110 Section 7.6.1. Versions 4.7.0 through 4.12.33 of the npm `hono` package are affected; the function correctly removed well-known hop-by-hop headers but neglected to parse and remove the additional headers the origin declared as connection-scoped via `Connection`. No public exploit has been identified at time of analysis and exploitation is conditional on non-default origin behavior, limiting real-world priority despite the network-reachable attack vector.
Improper behavioral workflow enforcement in WonderTrader up to version 0.9.9 allows remote attackers to manipulate the FID_JYLB argument in the TraderDD::queryTrades function, causing the application to execute trading query operations in an unintended sequence. Per the CVSS 4.0 vector (AV:N/AC:H/PR:N/UI:N), exploitation is network-reachable without authentication but requires high attack complexity, limiting opportunistic abuse. A proof-of-concept exploit is publicly available (E:P), and the vendor did not respond to responsible disclosure, leaving no patch available at time of analysis.
Root password hash exposure in ZTE F689 firmware allows a physical attacker to extract credential material stored in cleartext or weakly protected form within the device firmware image. The F689 is an ISP-deployed broadband gateway, and this vulnerability (CWE-916) means the hash protecting root-level access uses insufficient computational effort, making offline cracking feasible once extracted. No public exploit code has been identified and no active exploitation is confirmed; however, successful hash recovery grants an attacker a path to persistent root-level device compromise.
OS command injection in HKUDS nanobot 0.2.0 and 0.2.1 allows a low-privileged remote attacker to bypass the configured shell command allowlist by submitting chained shell commands using operators such as `&&`, `||`, `;`, `|`, or `&`. The `ExecTool._guard_command` method in `shell.py` validated the entire raw command string as a single unit against allow patterns, meaning only the first chained segment needed to match - subsequent malicious segments executed unchecked. A public proof-of-concept exploit exists; the vulnerability is not listed in CISA KEV but the CVSS 4.0 E:P modifier confirms publicly available exploit code elevates urgency beyond the base score of 5.3.
Unrestricted file upload in SourceCodester Photo Share Website 1.0 exposes the /social/ajax.php?action=save_upload endpoint to abuse via crafted img[] and imgName[] POST parameters, allowing authenticated low-privileged users to upload arbitrary file types - including PHP webshells. If the upload directory permits server-side script execution, this escalates to remote code execution on the host. A public proof-of-concept exploit exists on GitHub; the vulnerability is not listed in the CISA KEV catalog.
Use of an uninitialized variable in WonderTrader up to version 0.9.9 allows a remote low-privileged attacker to cause limited availability disruption by manipulating the m_offsetType argument within the TraderATP Cash Trade Conversion component. The flaw resides in src/Includes/WTSTradeDef.hpp and is exploitable remotely, with a public proof-of-concept available via a Feishu document. No patch has been released and the vendor has not responded to disclosure - exploitation conditions are low complexity once authentication is satisfied, making this a persistent unmitigated risk for deployments using the ATP cash trading interface.
Server-side request forgery in HKUDS nanobot up to 0.2.1 allows authenticated remote attackers to make the server issue HTTP requests to arbitrary internal endpoints by manipulating URLs returned by AI image generation providers. The `_download_image_data_url` function in `nanobot/providers/image_generation.py` fetched provider-returned image URLs directly via `client.get(url)` without routing through the existing `validate_url_target` SSRF guard in `security/network.py`, leaving cloud metadata endpoints (e.g., 169.254.169.254), RFC1918 ranges, and internal services reachable. A public POC exploit exists; a patch is available on the main branch as PR #5095 and is planned for release in v0.3.1. No CISA KEV listing; exploitation is not confirmed at mass scale.
Behavioral workflow enforcement bypass in WonderTrader up to version 0.9.9 allows remote low-privileged attackers to manipulate the getUndoneQty argument within the _undone_qty function of the Pending Order Handler component (src/WtCore/TraderAdapter.h), resulting in unauthorized integrity modification of pending order state. The CVSS 4.0 vector confirms network-accessible exploitation with low privilege requirements and no user interaction. A publicly available proof-of-concept exploit exists (E:P), though no CISA KEV listing is recorded. The vendor was notified prior to disclosure and did not respond, meaning no official patch or advisory has been released.
MCP resource and prompt wrappers in HKUDS nanobot 0.2.0 and 0.2.1 bypass operator-defined `enabledTools` access restrictions, allowing MCP servers to register capabilities beyond the intended scope. Operators who configure a specific tool allowlist to restrict what an MCP server can contribute to a nanobot agent find that resources and prompts from that server are registered unconditionally, exposing capabilities that were explicitly intended to be blocked. A public proof-of-concept exploit exists; this CVE is not in the CISA KEV catalog, so confirmed widespread active exploitation is not established.
Cross-site scripting in SourceCodester Photo Share Website 1.0 allows a low-privileged authenticated attacker to inject malicious JavaScript via the `content` parameter of the comment input box at `/social/ajax.php?action=save_upload`, executing arbitrary script in the browsers of users who subsequently view the injected content. The CVSS 4.0 vector (PR:L/UI:P) confirms exploitation requires an authenticated session and victim interaction, constraining opportunistic mass exploitation. No public exploit identified at time of analysis is incorrect here - publicly available exploit code exists, referenced via a GitHub issue (https://github.com/Andyov0/cve/issues/1), meaningfully elevating near-term risk for any reachable deployment.
Stored or reflected cross-site scripting in SourceCodester Photo Share Website 1.0 enables remote authenticated attackers to inject arbitrary JavaScript via the Comment argument at /social/index.php?page=home, executing in the browsers of other users who view the affected page. A publicly available proof-of-concept exploit exists on GitHub (gensnext/cve/issues/1), though the vulnerability carries a low CVSS 4.0 score of 2.0 and is not listed in the CISA KEV catalog, indicating no confirmed widespread active exploitation. Practical impact is limited to session hijacking, phishing redirection, or DOM manipulation within the application context of other authenticated users.
Stored cross-site scripting in PHPGurukul Company Visitor Management System 1.0 allows a high-privileged authenticated attacker to inject malicious JavaScript via the `fullname` parameter in `/manage-newvisitors.php`, executing arbitrary scripts in the browsers of other users who view the affected page. The CVSS 4.0 score of 4.8 (Medium) reflects the high privilege prerequisite (PR:H) and required passive victim interaction (UI:P), substantially limiting real-world impact. A public proof-of-concept exploit exists (GitHub), though no active exploitation has been confirmed by CISA KEV.
Login-shell environment variable leakage in HKUDS nanobot up to 0.2.1 exposes sensitive environment variables to local low-privilege users during agent-driven shell command execution. The `ExecTool._prepare_command` function in `shell.py` defaulted to invoking bash/zsh with the `-l` (login shell) flag, causing system and user profile startup scripts to be sourced and potentially re-exporting credentials, API keys, or tokens into the execution environment where they could be read by the invoking user. A publicly available proof-of-concept exploit exists; no active exploitation has been confirmed (not listed in CISA KEV). The vendor-released fix in v0.3.0 changes the default from login-shell to non-login-shell execution.
Use-after-free in Zephyr RTOS kernel queue peek operations (z_impl_k_queue_peek_head and z_impl_k_queue_peek_tail in kernel/queue.c) allows a local low-privileged thread to obtain a dangling pointer to freed heap memory by winning a race against a concurrent k_queue_get call on a shared queue. Successful exploitation - requiring SMP or preemptive concurrency - can leak a stale heap pointer or crash the system when the returned dangling pointer is subsequently consumed as a live buffer in the net_buf, Bluetooth, USB, or networking subsystems. No public exploit exists and no active exploitation is confirmed; CVSS scores this at 3.6 (Low), accurately reflecting the narrow race window and local-only attack surface.
Symlink following in File::Rotate::Simple before v0.4.0 allows a local low-privileged attacker to create arbitrary files with the permissions of the log-rotating process. When the non-default `touch` option is enabled and the log file path is a dangling symlink, the rotate method skips rotation (because the symlink target does not exist) but still invokes touch, which resolves the symlink and materializes the target at an attacker-controlled path. No public exploit code exists and EPSS is 0.17% (7th percentile), indicating negligible current exploitation interest.
Stale permission enforcement in Frappe's Document Follow subsystem allows authenticated users whose document access has been revoked or downgraded to continue receiving document-level data via scheduled email notifications. Affecting all Frappe v15 releases prior to 15.112.0 and v16 releases prior to 16.23.0, the flaw arises because notification generation never re-evaluates whether recipients still hold current read permission on subscribed documents at dispatch time. No public exploit code exists and no active exploitation has been confirmed; the vendor-assigned CVSS 4.0 score of 2.3 accurately reflects the narrow exposure window and the prerequisite that access revocation must have already occurred before the notification job fires.
CRLF injection in libvirt's virtual network driver enables privileged users holding network-definition rights to inject arbitrary dnsmasq configuration directives - including dhcp-script - into the generated dnsmasq config file, ultimately achieving arbitrary command execution as root on the hypervisor host. The root cause is the network XML parser's failure to strip newline characters from DNS TXT record value attributes and SRV record domain/target attributes before writing them verbatim into the dnsmasq configuration. No public exploit has been identified at time of analysis, though an upstream fix commit exists in the libvirt GitLab repository; the official CVSS score of 2.3 appears to critically understate the impact relative to the described root code execution primitive.
Stored cross-site scripting in Domoticz's mobile dashboard (all versions prior to 2026.3) allows any authenticated API user to inject persistent JavaScript payloads via Text or Alert subtype device values. The ng-bind-html rendering path applies only an nl2br() newline transform with no HTML escaping, meaning stored payloads execute in every administrator's browser that loads the mobile dashboard. Successful exploitation enables session cookie theft and account takeover; no active exploitation is confirmed (not in CISA KEV) and no public exploit code has been identified at time of analysis.
Heap-based buffer overflow in MZ Automation libiec61850 up to version 1.6.1 allows an attacker on the adjacent IEC 61850 SV network segment to corrupt process memory by injecting malformed Sampled Values ASDU packets into the SVReceiver. The commit diff confirms that the parseASDU function in SVReceiver_stopThreadless assigned heap pointers for ASDU fields-svId, datSet, SmpCnt, ConfRev, RefrTm, SmpSynch, SmpRate, and SmpMod-without validating their lengths, enabling overflow via crafted field sizes. A public exploit has been released (E:P per CVSS 4.0 supplemental data); this vulnerability is not currently listed in CISA KEV but poses targeted risk to IEC 61850 substation automation and power grid environments.