Sandbox escape in vxcontrol PentAGI up to 2.1.0 exposes a privilege boundary weakness in the Tool Management Protocol Handler, rooted in the pentester prompt template file. Remote unauthenticated attackers who can craft sufficiently complex inputs may bypass sandbox constraints on the AI agent's tool execution context, achieving limited confidentiality, integrity, and availability impact against the vulnerable system. A public proof-of-concept exists on GitHub; no patch has been released and the vendor has not responded to disclosure - leaving all deployments on versions up to 2.1.0 without a vendor-sanctioned fix.
Insecure Direct Object Reference (IDOR) in the Tag, Category, and Taxonomy Manager WordPress plugin before 3.51.0 allows authenticated contributors to disclose content from private or draft posts they do not own. The plugin fails to verify post ownership before processing a referenced post ID and returning derived data, enabling a contributor to enumerate and extract information from restricted content authored by other users. No public exploitation in the wild has been confirmed (no CISA KEV listing), though WPScan has published a publicly available proof-of-concept.
Unauthorized revenue data disclosure in the Classified Listing WordPress plugin (before 5.4.4) allows any authenticated contributor-level user to read daily store revenue figures restricted to administrators and report managers. The flaw stems from a missing capability check (CWE-862) on an AJAX action that aggregates store revenue totals, exposing potentially sensitive business intelligence to lower-privileged users. No public exploit identified at time of analysis per KEV, though a publicly available exploit exists per WPScan reporting; the EPSS score of 0.13% (3rd percentile) suggests limited real-world exploitation interest despite code availability.
The Classified Listing WordPress plugin before version 5.4.4 exposes private, draft, and pending post content to authenticated contributors due to a missing authorization and ownership check on an AJAX endpoint. Any user with contributor-level access or higher can read arbitrary post content belonging to other users, including administrators' unpublished material. A publicly available exploit exists via WPScan, though EPSS at 0.13% (3rd percentile) and the absence of a CISA KEV listing indicate active exploitation is not confirmed at time of analysis.
Path traversal in danpros HTMLy up to version 3.1.1 permits authenticated remote attackers to manipulate arbitrary file paths via the `oldfile` parameter in the Admin Content Endpoint (`/system/admin/admin.php`), enabling unauthorized file read or write operations outside intended directories. Exploitation requires low-privilege administrative access, and a public proof-of-concept exploit has been released on GitHub. The vendor was notified prior to disclosure but did not respond, meaning no patch has been issued.
Authorization bypass in jeequan jeepay up to version 3.2.9 allows authenticated low-privilege users to circumvent access controls enforced by the PreAuthorize Handler within WebSecurityConfig, specifically targeting the SysLogController component. The flaw stems from CWE-639 (Authorization Bypass Through User-Controlled Key), enabling manipulation of access-control keys to reach system log data or restricted administrative functionality beyond the user's granted permissions. A public proof-of-concept exploit exists; no vendor response has been received and no patch has been confirmed, elevating remediation urgency despite the moderate CVSS score.
Server-side template injection (SSTI) in Dify's Jinja2 Handler (versions ≤1.14.2) allows remote, low-privileged authenticated attackers to inject malicious Jinja2 template expressions through the code execution subsystem, resulting in information disclosure and potential further impact. A public proof-of-concept exploit is available on GitHub, materially lowering the bar for exploitation. The vendor did not respond to responsible disclosure, meaning no patch has been released and no official mitigation guidance exists.
Second-order SQL injection in osCommerce 4.14.63493 allows remote attackers with high-privilege (administrator-level) backend access to manipulate the `email_templates_key` parameter in the Email Template Configuration component, enabling database extraction and modification. Public exploit code is confirmed available via an anonymous repository hosting a detailed second-order SQLi report. No vendor patch exists - the vendor did not respond to responsible disclosure - raising operational risk for any organization still running this version despite the PR:H constraint.
Path traversal in Blix Email Blue Mail Calendar App 2.2.305 exposes the device filesystem to local attackers via the react-native-receive-sharing-intent component. The functions FileDirectory.getDataColumn and FileDirectory.getFileFromUri fail to sanitize the _display_name argument, allowing traversal sequences to escape the app's intended directory scope and access or overwrite arbitrary files reachable by the app's process. No active exploitation is confirmed in CISA KEV, though a public proof-of-concept is available on GitHub and the vendor did not respond to coordinated disclosure, leaving no official patch available.
Reachable assertion in llama.cpp's Jinja Minja Template Parser crashes the process when processing a maliciously crafted template input. The flaw, located in common/jinja/parser.cpp, is triggered locally by a low-privileged user supplying the input sequence {{9|9|{ to the template engine, causing an unhandled assertion failure and denial of service. A public proof-of-concept exploit has been disclosed; the project was notified via GitHub issue #25282 but has not issued a patch or response as of this writing.
Improper export of the DialerActivity component in textPlus Text Message and Call App (Android, versions up to 8.3.5) allows any co-installed application on the same device to invoke the com.gogii.textplus dialer component without authorization. Classified under CWE-926, this flaw permits a locally-present malicious app to trigger dialer functionality, leading to limited information disclosure and unauthorized interaction with call-related features. No public KEV listing confirms active exploitation, but a proof-of-concept is publicly available on GitHub, lowering the technical barrier for abuse.
Stored cross-site scripting in OpenAkita's File Upload API (/api/upload) allows remote attackers to inject persistent malicious scripts by manipulating file-related arguments, affecting all versions through 1.27.12. The Notion-hosted proof-of-concept confirms this is a storage-type XSS, meaning the payload persists server-side and executes in victims' browsers upon subsequent access to the affected content. A public exploit exists and the vendor did not respond to responsible disclosure, leaving no official patch; exploitation is constrained by high attack complexity (AC:H) and required user interaction (UI:P), reflected in the low CVSS 4.0 score of 2.3.
Cleartext storage of personally identifiable information in Meesho Online Shopping App (Android, versions through 20260607) exposes user_id, phone number, email address, and name in the com.meesho.supply component without encryption. Exploitation requires physical access to the device and a degree of local privilege, making this a low-severity, localized threat rather than a remote or mass-exploitation risk. A proof-of-concept is publicly available via GitHub and Google Drive, though no active exploitation has been confirmed by CISA KEV.
Incomplete cleanup of cached print job data in Sharp and Toshiba Tec multifunction printers exposes previously printed documents to subsequent physical users of the same device. Affecting all versions per CPE data (cpe:2.3:a:sharp_corporation:sharp_mfps and cpe:2.3:a:toshiba_tec_corporation:toshiba_tec_mfps), the flaw (CWE-459) means internal caches are not wiped after a print job completes, allowing the next person who physically accesses the device to potentially retrieve another user's data. No public exploit code exists and the vulnerability has not been added to CISA KEV, consistent with the low CVSS score of 2.4 and a physical-access attack vector.
Authorization bypass in Apache NiFi 2.0.0-2.10.0 allows authenticated users holding write access to one Parameter Context to delete Assets belonging to a separate, restricted Parameter Context they are not authorized to modify. The REST API's asset deletion endpoint validated permissions against the caller-supplied Parameter Context Identifier without confirming that the targeted Asset Identifier is actually recorded as belonging to that context - a broken object-level authorization (BOLA/IDOR) flaw. No public exploit code exists and the vulnerability is absent from the CISA KEV catalog; impact is confined to multi-tenant NiFi deployments with differentiated per-context authorization policies.
Insufficient permission checks in Misskey's Direct Messages (Chat) feature expose DM data points to authenticated users who lack authorization to access them, across versions 2025.3.2 through 2026.5.3. The root cause is CWE-639 (Authorization Bypass Through User-Controlled Key), where the server fails to validate account permissions before returning DM-related data via API. No public exploit or CISA KEV listing exists; the vendor patched this in version 2026.5.4 alongside four additional CVEs.
OS command injection in the Wavlink WL-NU516U1 router's admin password handler allows a remote, low-privileged authenticated attacker to execute arbitrary operating system commands via the set_sys_adm function in adm.cgi. The affected firmware version is 708c073-mt7628, and a public proof-of-concept exploit has been disclosed on GitHub. No CISA KEV listing exists, but the vendor has responded promptly and released a patched firmware dated 2026-07-13.
Path traversal in danpros HTMLy up to version 3.1.1 enables remote authenticated attackers to delete arbitrary files via the Delete Username endpoint. The PHP `unlink()` function in `/system/htmly.php` accepts a user-controlled `File` argument without path sanitization, allowing traversal outside the intended directory. A public proof-of-concept exploit exists (CVSS 4.0 E:P); no vendor patch is available as the vendor was unresponsive to disclosure, leaving installations permanently exposed without manual mitigation.