Command Injection
Monthly
OS command injection in the FILESystem:MOUNT:DRIve SCPI command handler of Tektronix 3 Series MDO oscilloscopes running firmware before 1.21.1 allows authenticated, low-privileged attackers who can reach the instrument's remote-control interface to execute arbitrary operating system commands with elevated privileges (CVSS 8.8, CWE-78, PR:L/UI:N). Any 3 Series MDO that exposes its SCPI control interface (LAN/VXI-11 or raw socket) to a lab or engineering network is in scope, and the attack requires no user interaction. No public exploit code and no CISA KEV listing were identified at time of analysis, and no EPSS score was supplied in the available data, so the realistic risk is concentrated on network-exposed instruments rather than mass exploitation.
OS command injection in ZoneMinder allows an authenticated user holding monitor-create or monitor-edit permission to execute arbitrary commands as the web server user by saving a Local-type monitor whose Device Path field contains shell metacharacters. The malicious stored Device value is later interpolated into shell commands in both Perl (qx()/backticks via zmdc.pl control paths and zmpkg.pl start/restart) and PHP (zmc daemon control helpers that relied on bypassable escapeshellcmd) when the monitor control or startup path runs. The issue is fixed in 1.39.2, 1.38.2, and 1.36.40, which reject any Device value not matching a strict /dev/ path pattern; no public exploit code has been identified at time of analysis, and exploitation is not confirmed actively exploited (KEV).
Remote command execution in ZoneMinder prior to 1.38.3 allows an authenticated user with monitor-create or monitor-edit rights to store a monitor Name containing shell metacharacters that break out of manual single-quote wrapping when a user with Events View permission exports that monitor's events, executing arbitrary commands as the web server user. Exploitation requires two authenticated roles and victim interaction (CVSS vector PR:L/UI:R), so this is not a default-reachable or mass-exploitable issue; no public exploit identified at time of analysis, and a vendor-released patch is available in 1.38.3.
Authenticated OS command injection in ZoneMinder versions prior to 1.38.3 allows a low-privilege user holding View Events or View Snapshots permission to run arbitrary operating system commands as the web server user. The flaw lives in exportEvents() in web/skins/classic/includes/export_functions.php, where the attacker-controlled exportFile request parameter is appended to a tar or zip command executed via exec() without shell escaping, so a crafted value can break out of the trailing directory argument. Exploitation requires a valid session (CVSS PR:L), a valid CSRF token for the export request, and the exportDetail option enabled; no public exploit code has been identified at time of analysis, and there is no confirmed active exploitation (CISA KEV), though the vendor's own GHSA advisory and fix commits confirm the issue is real and the EPSS-style risk is elevated by the low privilege threshold and high confidentiality, integrity, and availability impact (CVSS 8.8).
OS command injection in SparkyFitness versions prior to 1.6.0 allows an authenticated global administrator to execute arbitrary operating-system commands on the server by uploading a restore archive with a filename containing shell metacharacters. The vulnerability resides in the api/admin/backup/restore endpoint, which improperly interpolates the user-supplied filename into a shell command. Exploitation requires the highest-level application privilege (global admin), and no public exploit code has been identified at time of analysis. The issue is fixed in version 1.6.0.
Remote code execution in the Theme-one The Grid WordPress plugin through version 2.8.1 allows network attackers to inject code or commands via the plugin's grid-generation code path, with the supplied CVSS:3.1 vector indicating AV:N/AC:L/PR:N/UI:N (unauthenticated and no user interaction per the vector, though the exact endpoint and authentication prerequisite should be verified in the Patchstack advisory). The provided vector rates this as 10.0 Critical with full confidentiality, integrity, and availability impact, while our independent assessment adjusts scope from Changed to Unchanged but still considers impact High. No public exploit identified at time of analysis, and no KEV or EPSS data was provided.
Authenticated OS command injection in the WP Photo Album Plus WordPress plugin (all versions up to and including 9.3.02.002) lets any WordPress user granted the plugin-defined wppa_upload capability execute arbitrary operating-system commands as the web-server user, achieving full server-side code execution where ImageMagick is the active image backend. Rated CVSS 8.8 (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), the flaw is reachable through the update-photo AJAX action's value parameter, and no public exploit code was identified at time of analysis; the bug is not in CISA KEV. Because wppa_upload is a plugin-level privilege that administrators can delegate to lower-privileged roles, the effective attacker pool can extend well beyond site administrators on misconfigured sites.
Command injection (CWE-78) and stored XSS in Wheels CLI 4.0.0 through 4.1.1 allow an attacker who can influence inputs processed by the tool to run arbitrary shell commands as the SSH deploy user on deploy hosts and on the CI runner during builds. Deploy release/version/rollback values, package removal/destroy and extraction paths, and generator names are the primary injection surfaces, with unencoded scaffold 'show' views (since 4.1.0) and generated admin views permitting stored XSS; trigger inputs can come from CI variables, Git branch names, downloaded packages, or an AI agent using the CLI's MCP tools. The flaw is fixed in 4.1.2, and existing affected generated views also require a documented manual edit; no public exploit code has been identified at time of analysis.
Command injection in pfSense CE 2.7.0 through 2.9.0 allows authenticated users who hold only the Captive Portal HA privilege to execute commands as root by manipulating backwardsyncip to point at an attacker-controlled XMLRPC server and returning connected_users entries with shell metacharacters in ip or mac fields. Exploitation is privilege-gated: it requires the specific Captive Portal HA WebGUI permission, use of Captive Portal high-availability sync, and the ability to influence the synced connected_users data. Publicly available exploit code exists, but no CISA KEV confirmed active exploitation is identified; this is a credible privilege-gated priority rather than a mass-exploitation emergency.
Command injection in Netgate pfSense CE 2.4.5 through 2.9.0 lets an authenticated WebGUI user holding the Advanced Admin Access privilege execute arbitrary commands as root by placing newline or $(...) sequences in the Login Protection (sshguard) threshold, block time, or detection time fields, which are written unvalidated to sshguard.conf and sourced as root when the service restarts. The flaw is privilege-gated rather than mass-exploitable: the CVSS vector specifies PR:H, so only already-administrative accounts can trigger it, with no user interaction required beyond the attacker's own session. Publicly available exploit code exists (referenced in the CVE record), and the root cause is CWE-78 OS command injection.
Authenticated VPN users can execute arbitrary commands as root on pfSense CE through 2.9.0 by exploiting an OS command injection in the OpenVPN connect/disconnect script, openvpn.connect_async.sh. The attack requires a valid VPN account authenticated via RADIUS or LDAP and a username containing single quotes and shell metacharacters that break out of an eval'd pfctl anchor name. Publicly available exploit code exists (reported by VulnCheck), though no active exploitation has been confirmed by CISA KEV. This is a credible high-priority issue for any pfSense CE deployment that uses OpenVPN with external authentication.
OS command injection in Netgate pfSense CE 2.7.0 through 2.9.0 lets an authenticated web GUI user who holds the Traffic Shaper Limiters privilege execute arbitrary commands as root, turning a narrowly delegated configuration permission into full appliance compromise. The flaw is triggered when an operator edits a limiter and injects a crafted hidden 'number' value and then deletes that limiter, causing delete_queue() to pass the unquoted, attacker-controlled value to dnctl via /bin/sh running as root. Publicly available exploit code exists (VulnCheck advisory plus Abraxas Labs research and a PoC repository), and while the issue is not listed in CISA KEV as actively exploited, exploitation is gated on prior authenticated access with the Limiters privilege (PR:H) and on the attacker completing the limiter edit-then-delete workflow.
Command injection in Netgate pfSense CE through 2.9.0 allows an authenticated user with Virtual IP (CARP) edit privileges to execute commands as root by storing a malicious VHID value that is later interpolated into ifconfig calls or CARP status rendering. Publicly available exploit code exists, and the independent assessment rates this as a privilege-gated high-impact issue rather than an unauthenticated remote compromise; CISA KEV does not confirm active exploitation in the supplied data. Affected deployments are those where administrators or delegated operators can create or modify CARP VIPs, narrowing realistic risk to insiders or attackers who already hold privileged firewall UI access.
OS command injection in Netgate pfSense CE (through 2.9.0) allows remote attackers holding an authenticated web-GUI account with gateway edit privileges to execute arbitrary commands as root by planting shell metacharacters in the zone index of an IPv6 link-local gateway Monitor IP, which start_dpinger() passes unescaped into the dpinger shell command when gateway monitoring is applied. Exploitation requires high privileges (CVSS 4.0 PR:H) and no victim interaction (UI:N), but yields high confidentiality, integrity and availability impact on the firewall itself; publicly available exploit code exists (abraxas/pfsense-gw-ipv6-zone) with no confirmed active exploitation reported. This is a genuine but access-gated priority: unauthenticated or low-privileged users cannot reach the vulnerable gateway-edit path, so exposure is limited to deployments where such configuration rights are delegated.
Command injection in Netgate pfSense CE through 2.9.0 allows remote authenticated users with limited 'Wireless Edit' or 'Interface Assignments' GUI privileges to execute arbitrary commands as root. The flaw is triggered by supplying a crafted wireless parent interface name such as ath0;<cmd> through interfaces_wireless_edit.php or interfaces_assign.php, which interface_wireless_clone() interpolates unescaped into an ifconfig shell command. Publicly available exploit code exists, and the assessed impact is high for confidentiality, integrity, and availability, though exploitation requires valid credentials with one of those specific privileges and access to the management interface.
OS command injection in Netgate pfSense CE 2.7.0 through 2.9.0 allows an authenticated web-interface user who holds the QinQ edit privilege to execute arbitrary commands as root. The attacker saves a crafted parent interface ('if') value and then re-edits the QinQ entry, causing that unvalidated value to be interpolated into an ifconfig shell command; publicly available exploit code exists, and no confirmed active exploitation (CISA KEV) has been reported. The principal constraint is the high privilege requirement - only operators already trusted with interface configuration can reach the vulnerable code path - and risk is further reduced where the management UI is not exposed to untrusted networks or lower-privileged users.
OS command injection in Unitronics UniLogic (versions prior to 1.41.329) and the UniStream Pro B10/C10 controllers it programs allows an authenticated low-privilege attacker who can reach a vulnerable input field to execute arbitrary operating-system commands on the affected device, resulting in full compromise of confidentiality and integrity plus partial availability impact. The flaw is network-reachable with low attack complexity and no user interaction, but requires valid credentials (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), which limits practical exposure to devices whose programming or web interface is reachable from an untrusted network segment. No public exploit code or confirmed active exploitation was identified at time of analysis; the issue was reported by Israel's National Cyber Directorate (INCD).
OS command injection in Wavlink WL-WN530BE2-A firmware up to M30BE2_V250604 lets remote attackers reach the api.cgi login handler (api=info&client=app, function sub_12580) and inject shell metacharacters through the value the firmware treats as the client address (REMOTE_ADDR), potentially yielding arbitrary command execution with root-level privileges on the device. The vendor has confirmed the finding and shipped a fix in firmware 2026-09-24-841a51c6d5_tclinux.bin, which replaces the system() call with execvp and length-limits/null-checks the client address; publicly available exploit code exists (a GitHub PoC), though the weakness is not confirmed as actively exploited (CISA KEV). Exploitation is not as trivial as the raw CVSS 4.0 score of 9.3 implies: REMOTE_ADDR is conventionally derived from the TCP source address, so a real-world attacker likely needs the web server to trust a client-supplied address header or needs an on-path/proxy position, which materially narrows the population of reachable targets despite PR:N in the published vector.
Unauthenticated remote command execution in QNAP HybridDesk Station (HD Station) lets network attackers run arbitrary operating system commands on affected QNAP NAS devices that have the optional desktop/multimedia add-on installed. The root cause is an OS command injection (CWE-78) rated 9.3 under CVSS v4.0 (AV:N/AC:L/PR:N/UI:N), so no credentials, user interaction, or special preconditions are required; the vendor has fixed it in HybridDesk Station 4.2.23 (4.2 branch) and 4.3.4 (4.3 branch). No public exploit code or confirmed active exploitation has been identified at time of analysis, but real-world risk is governed by whether HD Station is installed and whether its interface is reachable from the attacker's network position (LAN versus internet-facing), as the specific vulnerable feature and input parameter have not been disclosed.
OS command injection in the pfSense-pkg-frr package (2.0.2_1 and source revision 2.1.2_1) lets an authenticated WebGUI user holding only the low-privilege page-status-frr role execute arbitrary commands as root on the firewall. The flaw sits in the FRR status filter on status_frr.php, where a crafted filter value escapes a single-quoted sed expression that concatenates escapeshellarg() output before being evaluated through a popen() pipeline. Exploitation requires the optional FRR package to be installed plus a valid authenticated session with the page-status-frr privilege (CVSS 3.1: AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, score 8.8), so it is a post-authentication privilege escalation rather than a pre-auth compromise; there is no public exploit identified at time of analysis, and no CISA KEV listing. Updated FRR packages resolve the issue for pfSense Plus 26.07/26.03.1 and CE 2.9.0/2.8.1.
Arbitrary OS command execution on hosts running MoviePilot through 3.1.5 is achievable when the product's LLM agent ingests attacker-controlled content via browse_webpage, search_web, or torrent metadata parsing and is induced to call the unrestricted ExecuteCommandTool, which runs shell commands through asyncio.create_subprocess_shell under the admin session's is_admin context. The flaw is a prompt-injection-to-RCE chain (CWE-1427) that bypasses the validate_command_safety denylist by setting confirm_dangerous=true; the attacker needs no MoviePilot credentials (CVSS PR:N), but an administrator must interact with the poisoned content (UI:P in CVSS:4.0, UI:R in the assessed CVSS:3.1) and the model must be successfully coerced, so the issue is conditionally gated rather than a mass-exploitation emergency. Severity is 8.7 in CVSS:4.0 and 8.7 in the independently assessed CVSS:3.1; no public exploit code and no CISA KEV entry are associated as of this analysis.
Authenticated attackers with high privileges on SonicWall Network Security Manager (NSM) on-prem can execute arbitrary OS commands via an OS command injection flaw (CWE-78), achieving remote code execution. The vulnerability affects NSM on-prem versions 4.3.1 and earlier, requires a privileged session (PR:H), and is assessed with high attack complexity (AC:H) due to unenumerated specific conditions. No public exploit code or active exploitation has been identified (CISA KEV status: none; SSVC exploitation: none), and the EPSS score is 0.54% (44th percentile), indicating a moderate-priority risk rather than an urgent emergency.
Command injection in the Markdown Preview Enhanced VS Code extension (which bundles the crossnote library) on Windows allows an attacker-controlled Markdown link to execute arbitrary commands with the privileges of the VS Code user. When a victim opens an untrusted Markdown file, previews it, and clicks a crafted external or unsupported-scheme link, the decoded link target is passed to a Windows file-opening helper that invokes child_process.exec with the start command, so embedded cmd.exe metacharacters are interpreted as command separators. The flaw affects versions prior to 0.8.28 (crossnote prior to 0.9.29), requires user interaction (opening the file and one link click), needs no workspace trust, no script execution, and no additional files, and earns a CVSS 3.1 base score of 8.8; no public exploit code was identified at time of analysis and the issue is not in CISA KEV.
Unauthenticated OS command injection in the D-Link DIR-868L router running firmware 2.01b05 lets remote attackers execute arbitrary shell commands on the device by sending a crafted SOAPACTION argument to the HNAP handler exposed through /htdocs/cgibin. The flaw is remotely reachable with no credentials, no user interaction, and no non-default configuration required (CVSS v4.0 base 9.3, assessed v3.1 as AV:N/AC:L/PR:N/UI:N/C:H/I:H/A:H), so any installation exposing the HNAP management interface is at risk; LAN-only deployments are largely confined to local-network exploitation, while devices with WAN-side remote management enabled are directly attackable from the internet. Publicly available exploit code exists (a detailed advisory and PoC are published on GitHub), which raises the practical urgency, though active exploitation is not confirmed by CISA KEV, and the product is an end-of-life model for which no vendor-released patch has been identified.
OS command injection in the phcd phone-home daemon of Juniper Networks Junos OS and Junos OS Evolved lets an adjacent, unauthenticated attacker who operates a rogue DHCP server execute arbitrary commands as root on affected EX Series, SRX Series, and QFX Series devices. Exploitation is not remotely routable: the attacker must share a Layer-2 broadcast domain with the target and the device must be actively running phone-home / zero-touch provisioning (ZTP), which is why an independent assessment rates the attack complexity higher than the vendor's 8.7 score despite the high confidentiality, integrity, and availability impact (PR:N/UI:N once those conditions hold). No public exploit code has been identified at time of analysis and the flaw is not confirmed actively exploited (no CISA KEV entry), though NVIDIA/ENISA EUVD tracking and a full Juniper advisory (JSA112187) with vendor patches indicate this is a genuinely high-priority, patchable issue.
Authenticated, low-privileged TACACS+ users on Juniper Networks Junos OS Evolved can escalate to root and execute arbitrary OS commands, but only when the periodic TACACS+ authorization-refresh feature is enabled - an OS command injection flaw (CWE-78) that requires both a valid low-privileged TACACS+ account and that specific configuration to be active. The vendor rates it 9.3, while our independent assessment scores it CVSS:3.1 9.0 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H), reflecting that exploitation is not possible by an anonymous remote attacker despite the vendor's PR:N vector. There is no public exploit identified at time of analysis, and the flaw does not trigger on devices using TACACS+ without authorization-refresh, or on devices using RADIUS or local authentication.
OS command injection in the Junos OS CLI lets an authenticated, low-privileged local user escalate to full root control of the device by passing unsanitized arguments to the 'request csds' operational command. Only Junos OS 24.4 (from 24.4R1 up to before 24.4R2-S4), 25.2 (before 25.2R2-S1), and 25.4 (before 25.4R1-S3) are affected; releases before 24.4R1 are explicitly unaffected, and the attack is strictly local with no network or unauthenticated path. There is no public exploit identified at time of analysis, and Juniper has released fixed versions (24.4R2-S4, 25.2R2-S1, 25.4R1-S3), making this a serious but access-gated issue rather than an internet-facing emergency.
Unauthenticated OS command injection in Juniper Networks Junos OS Evolved's bootzd zero-touch provisioning (ZTP) daemon allows adjacent attackers on the management LAN to execute arbitrary commands as root via a crafted DHCP response. The vulnerability has a CVSS 4.0 base score of 9.3 and requires the target device to be actively in ZTP mode; no public exploit code is identified at time of analysis.
Unauthenticated OS command injection in the bootzd zero-touch provisioning (ZTP) daemon of Juniper Networks Junos OS Evolved on PTX Series routers allows an attacker positioned on the management LAN to execute arbitrary commands as root by sending a crafted gRPC response to a device that is actively provisioning. The vulnerability carries a CVSS 3.1 base score of 8.8 (AV:A/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H), reflecting low attack complexity, no authentication or user interaction requirements, and full confidentiality, integrity, and availability impact; the adjacent-network attack vector (AV:A) means the device is not exploitable directly from the internet. No public exploit code has been identified at time of analysis, no CISA KEV listing or EPSS score was provided, and exposure is materially reduced where management traffic runs over a segmented or physically isolated out-of-band network and where devices are not in an active ZTP state.
mise (a polyglot dev tool manager for node, python, cmake, terraform, and others) before version 2026.7.14 lets a malicious local project inject interpreter arguments through an untrusted .mise.toml [settings] file, causing attacker-chosen commands to execute with the developer's privileges. In default non-safe mode, mise applies local project settings before trust approval, and the unix_default_file_shell_args, unix_default_inline_shell_args, windows_default_file_shell_args, and windows_default_inline_shell_args fields were missing the global_only marker, so they survived strip_local_only_settings. When a developer later runs a trusted task, postinstall directive, template exec, or hook that has no explicit shell, mise appends the legitimate command after the attacker-controlled interpreter arguments, and the attacker's command runs first despite the config never being trusted. Exploitation is local and requires user interaction (the victim must clone/open and run the affected task); there is no public exploit code and no confirmed active exploitation at time of analysis, and this is best characterized as a conditional local-escalation risk rather than a mass-exploitation threat. Fixed in 2026.7.14.
A never-ending loop condition in curl and libcurl 8.10.0 through 8.22.0 lets a malicious, compromised, or on-path-manipulated server hang the client by omitting an expected part of its response, producing availability-only impact (the affected transfer or process spins indefinitely) with no confidentiality or integrity effect. The curl project rates this flaw LOW severity, and it is one of twenty-four CVEs disclosed alongside the curl 8.23.0 release. Exploitation requires the victim to connect outbound to the attacker's endpoint, since curl is a client and cannot be attacked purely remotely against a listening service; the exploit is not confirmed as actively exploited (not in CISA KEV), no public exploit code has been identified at time of analysis, and no EPSS score was provided in the source data.
Stack exhaustion in libcurl's MIME handling can crash applications that construct deeply nested multipart structures: curl_mime_free() recurses once per nesting level, so a sufficiently deep curl_mime_subparts() hierarchy overflows the stack and aborts the process. Affected are libcurl/curl versions 7.56.0 through 8.22.0, fixed in 8.23.0; the curl project classifies this advisory as LOW severity, with the impact limited to denial of service (no code execution, no data disclosure). The assessed vector (CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L) reflects that exploitation is confined to libcurl-based applications that let untrusted input dictate MIME nesting depth, and there is no public exploit identified at time of analysis.
OS command injection in the thumbnail Python package through version 1.5 allows attackers who control a filename passed to generate_thumbnail() to execute arbitrary shell commands with the privileges of the Python process. The vulnerable code concatenates user-influenced values (input, output, width, height, quality, trim) directly into os.system() calls invoking ffmpeg, convert or unoconv, so shell metacharacters such as ';', backticks or $(...) in a filename break out of the intended command. The package is a library with no network listener of its own, so real-world reachability depends on the consuming application exposing untrusted filenames - classically a file-upload feature - and on ffmpeg/convert/unoconv being installed for the relevant code branch; where those conditions hold, no authentication or user interaction beyond submitting the malicious filename is required. An upstream fix exists in pull request #9, and no public exploit code has been identified at time of analysis.
OS command injection in Stills (shahkashani/stills), a Node.js media-processing library, before version 25.24.2 allows crafted media file names containing shell command-substitution syntax such as $(...) or backticks to be executed as operating system commands with the privileges of the Node.js process. The flaw is reachable whenever the library shells out to ffmpeg, ffprobe, gifsicle, or convert on an attacker-influenced path, and the CVSS 4.0 vector (9.3) describes an unauthenticated, low-complexity network attack (PR:N/UI:N) yielding high confidentiality, integrity, and availability impact; our independent assessment scores it CVSS:3.1 8.8/9.8-class (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) but cautions that real-world risk is materially lower in deployments where media file names are operator- or system-generated rather than user-supplied. No public exploit code or confirmed active exploitation was identified at time of analysis, and a vendor patch (25.24.2, via PRs #29/#30 and associated commits) is available.
OS command injection in @offactory/compress-fonts before 1.1.1 lets an attacker execute arbitrary shell commands with the privileges of the Node.js process by getting a font file with shell metacharacters in its filename into a directory that is being compressed. The supplied CVSS 4.0 vector indicates no authentication is required (PR:N) and high confidentiality, integrity, and availability impact to the vulnerable component, but the attack vector is local (AV:L) and requires a user or automated build to initiate the FontForge conversion (UI:P), so real-world impact is bounded by who can place files in the processed source directory and by the privileges under which the build runs. No public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV; a vendor fix is available in version 1.1.1, which replaces execSync with execFileSync.
OS command injection in Turnpress through 0.4.2 allows arbitrary shell commands to be executed with the victim user's privileges when the tool's convert or create operation processes a .docx file whose filename contains shell metacharacters. The flaw sits in convertDocxToHtml (src/converter.ts), which invokes pandoc through execa with shell:true, so the crafted filename is interpreted by the shell rather than treated as a literal argument; exploitation is local (AV:L) and requires the victim to run a convert/create command on the malicious file (UI:R), and no privileges are needed to craft the filename. Impact is high for confidentiality, integrity, and availability on the host running the conversion, but it is limited to the invoking user's privileges with no scope change, and no public exploit code or confirmed in-the-wild exploitation has been identified at time of analysis; vendor fix is available via an upstream pull request that removes shell:true.
OS command injection on Windows in Beganovich Snappdf versions 1.1.0 through 6.0.x lets remote attackers execute arbitrary operating-system commands as the PHP process by feeding cmd.exe metacharacters (such as '&') into the URL passed to setUrl(). Because the vulnerable code path is built only when the library runs on Windows, the issue is exploitable only when the integrating PHP application runs on Windows and passes attacker-influenced data into setUrl(); a fully static or trusted URL is not exploitable. The CVSS v4.0 vector (AV:N/PR:N/UI:N) indicates the injection can be triggered remotely without authentication, giving high confidentiality, integrity, and availability impact on the host, although no public exploit code or confirmed in-the-wild exploitation has been identified at time of analysis.
OS command injection in the Video-stitch Node.js library (ArsalanDotMe) before 2.0.0-beta.1 allows unauthenticated remote attackers to run arbitrary shell commands with the privileges of the Node.js process by supplying file names containing shell metacharacters to the concat(), cut(), or merge() functions. The vulnerable code interpolated those values unquoted into shell-executed ffmpeg command lines (shelljs.exec) in lib/videoconcat.js and lib/videocut.js, so exposure is realized whenever an embedding application routes untrusted data into the output/clip name or fileName parameters; applications that only pass trusted, internally generated file names are not exploitable. CVSS 4.0 rates this 9.3 (Critical) with PR:N and UI:N, and no public exploit code or confirmed active exploitation was identified at time of analysis; a fix is available in version 2.0.0-beta.1.
OS command injection in nodejs-whisper through 0.3.1 lets remote, unauthenticated attackers execute arbitrary operating-system commands with the privileges of the Node.js process by supplying a crafted audio file path. The vulnerable convertToWavType() function builds an ffmpeg command as a string, interpolating the caller-provided path between double quotes before handing it to shelljs exec(); a filename containing $() command substitution or a closing double quote breaks out of the quoted argument and is interpreted by the shell. An upstream fix exists in pull request #259, but no tagged patched release has been independently confirmed, and no public exploit code was identified at time of analysis - real-world exposure is limited to applications that pass attacker-influenced, unsanitized filenames into the transcription/convert entry points while ffmpeg is installed on the host.
OS command injection in the transcribe2texts Node.js library through version 1.0.6 lets unauthenticated remote attackers execute arbitrary shell commands with the privileges of the Node.js process by supplying crafted audioPath values to transcribeAudio(). The flaw is a library sink rather than a directly exposed service: it is exploitable only where an integrating application passes attacker-influenced data into audioPath, which is concatenated unescaped into a python3 command line and run through child_process.exec(); applications that hardcode or strictly validate that path are not exposed. The critical 9.3 (CVSS 4.0) severity reflects network-reachable, low-complexity, no-privilege, no-interaction exploitation with high confidentiality, integrity and availability impact to the host, and while an upstream fix exists, no public exploit code has been identified at time of analysis.
OS command injection in xhtml2pdf before 0.2.19 lets an attacker execute arbitrary OS commands as the invoking user when the pisa command-line tool is run with the -s/--start-viewer option against a file whose name contains a double quote plus shell metacharacters. The flaw lives in pisa.startViewer(), which interpolates an unescaped filename into os.system('open "%s"'); because the vulnerable path is only reached in viewer-launch mode, the default programmatic HTML-to-PDF conversion API is not affected. The vector is local and requires a user (or job) to actually launch the viewer, and no public exploit code has been identified at time of analysis - confirmed actively exploited status is not present in the available data.
OS command injection in Runnerty's @runnerty/executor-decompress package (all versions through 3.1.1) lets an attacker who can control the name of an archive file processed by a compress_file/decompress task execute arbitrary shell commands with the privileges of the Runnerty process. The package builds shell command strings from templates that interpolate the archive filename and then hands them to shelljs exec(), so a filename containing $() command substitution or unquoted shell metacharacters - for example one dropped into a watched or ingested directory - is interpreted by the shell instead of being treated as a literal path. Exploitation requires low-privileged access sufficient to name files in that directory (PR:L) and no user interaction; impact to confidentiality, integrity, and availability is high but is bounded to the Runnerty process's privilege level, and no public exploit code or confirmed active exploitation was identified at time of analysis.
OS command injection in the whisper-node npm package through version 1.1.1 allows remote, unauthenticated attackers to run arbitrary shell commands with the privileges of the embedding Node.js process whenever an application passes attacker-controlled data into whisper()'s filePath argument or the whisperOptions.language (and potentially modelPath) options. The vulnerable code path is the default in all versions up to and including 1.1.1, with no special configuration required, although real-world exposure is gated by how the library is used: whisper-node is a library, not a network service, so only applications that feed it untrusted input are remotely exploitable, and impact is bounded by the Node.js process privileges. No public exploit code has been identified at time of analysis, and there is no indication of confirmed active exploitation (CISA KEV); the upstream fix appears in GitHub PR #64 but a released patched version could not be independently confirmed from the available data.
OS command injection in @mborne/extract versions before 0.2.1 lets an attacker execute arbitrary shell commands with the Node.js process's privileges by supplying shell metacharacters in the archivePath or targetDir options passed to extract(). The flaw is unauthenticated (CVSS 4.0 vector AV:N/AC:L/AT:N/PR:N/UI:N) and scored 9.3, with full confidentiality, integrity, and availability impact to the host process, though it is a library with no standalone attack surface - it is only reachable when the integrating application forwards attacker-controlled values into one of the affected extractor paths (unzip, 7z, tar-bz2, or gz; the 7z path additionally requires the 7z binary). No public exploit code was identified at time of analysis, so exploitation is contingent on the host application's input handling rather than on any inherently exposed service.
OS command injection in doc-recog through 1.1.6 allows remote attackers to execute arbitrary commands with Node.js process privileges by supplying crafted document paths containing $(...) or backtick command substitution to the recog() or tag() functions, which flow to a shell-based pdftotext exec call in DocData._readPDF. The CVSS vector indicates no authentication (PR:N) and no user interaction (UI:N), but real-world exploitability hinges on whether a consuming application passes attacker-controlled paths into the library; if paths are hardcoded or validated downstream, the flaw is not reachable. No public exploit code has been identified at time of analysis, and the upstream fix (PR #2) switches to execFile with an argument array to eliminate shell interpretation.
OS command injection in buildebooks (wastemobile) versions before 0.2.2 lets a crafted EPUB file name or project folder name break out of the shell command string built in lib/kindle.js and execute arbitrary operating-system commands with the privileges of the user running the build. Exploitation is local rather than network-reachable: the victim must process an attacker-supplied book project with the Kindle output path enabled (kindle: true), which satisfies the user-interaction prerequisite, and no privileges are needed on the attacker's side per the assessed vector (PR:N). No public exploit code or confirmed in-the-wild activity has been identified at time of analysis, but the affected code path is straightforward (a naive string concatenation into child_process.exec) and a vendor fix is available in 0.2.2.
OS command injection in the Python library pypdftk (through version 0.5) lets attackers who control the pdf_path or out_file arguments of fill_form(), dump_data_fields(), or pdftk_cmd_util() execute arbitrary operating-system commands with the privileges of the hosting Python process. The three functions concatenate caller-supplied paths into a command string using '%s' formatting and pass it to subprocess.Popen with shell=True, so shell metacharacters embedded in a filename or path are interpreted by the shell rather than treated as literal arguments. CVSS v4.0 scores the flaw 9.3 (AV:N/AC:L/AT:N/PR:N/UI:N), and the independent assessment concurs with CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H, but real-world reachability is bounded by the calling application - pypdftk is a library, not a network service, so exploitation requires the embedding application to forward untrusted input into those parameters, and no public exploit code has been identified at time of analysis. No CISA KEV entry exists for this CVE, and a source-level fix is available from upstream via pull request 57, which replaces string command construction with argument lists.
OS command injection in webp-converter before 3.0.0 lets unauthenticated remote attackers execute arbitrary operating system commands as the Node.js process when a host application routes attacker-controlled file names or option strings into the library’s cwebp, dwebp, gwebp, or webpmux helper functions. The flaw carries a CVSS 4.0 score of 9.3 (with an independent CVSS 3.1 assessment of 9.8), but because webp-converter is a library rather than a listening service, real-world risk depends on whether the integrating application actually passes untrusted input into the vulnerable code path; applications that use only server-controlled, fixed paths and options are not exposed. No public exploit code has been identified at the time of analysis, and the vendor has released version 3.0.0 to fix the issue.
OS command injection in the npm package git-diff (danday74/git-diff) before 2.0.7 lets unauthenticated attackers execute arbitrary shell commands with the privileges of the Node.js process by passing crafted document strings into the gitDiff API. The library hashed each input with a shell command of the form `printf "..." | git hash-object -w --stdin` evaluated through shelljs.exec, so payloads containing $(cmd) or backtick substitutions survived the JSON.stringify escaping and were executed. The vendor has published an upstream fix that removes the shell invocation entirely, and no public exploit code has been identified at time of analysis; the assessed CVSS 4.0 base score is 9.3 (Critical) for maximal confidentiality, integrity and availability impact to the vulnerable system.
Unauthenticated OS command injection in TR7 Cyber Defense Inc.'s TR7 ASP appliance (both virtual and physical deployments) allows any attacker who can reach the appliance's exposed service to execute operating-system commands without supplying credentials or user interaction. The flaw is rooted in a missing authentication check for a critical function (CWE-306) and carries a CVSS 3.1 base score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H), meaning full loss of confidentiality, integrity, and availability of the appliance if exploited. All TR7 ASP installations running 1.0.40.236 up to (but not including) 1.0.1554.2331, or 1.6.42.239 up to (but not including) 1.6.1554.2331, are affected; exposure is the primary gating factor, since appliances segmented onto trusted management VLANs are far harder to reach. No public exploit code and no CISA KEV entry have been identified at time of analysis, and no EPSS score was provided with the source data.
Remote unauthenticated OS command injection in TR7 ASP allows an attacker who can reach the appliance over the network to execute arbitrary operating system commands with the privileges of the vulnerable service, yielding full loss of confidentiality, integrity, and availability. Affected builds are TR7 ASP from 1.0.40.236 before 1.0.1554.2331 and from 1.6.42.239 before 1.6.1554.2331, covering both virtual-appliance and physical-appliance deployments. The flaw carries a CVSS 3.1 base score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) per the assessed vector, meaning no authentication, no user interaction, and low attack complexity; no public exploit code has been identified at time of analysis and the issue is not listed as confirmed actively exploited (CISA KEV). The primary limiting factors are version and network exposure: only the specified build ranges are affected and the appliance must be reachable by the attacker.
OS command injection in the Koha integrated library system lets an authenticated staff account holding the tools => schedule_tasks permission execute arbitrary operating-system commands as the Koha web/application user. The flaw lives in the tools/scheduler.pl handler: an op=cud-delete request passes the unvalidated jobid parameter into C4::Scheduler::remove_at_job, which hands it to Schedule::At::remove, where the value is spliced into an atrm command string and run through a single-argument system() call that invokes the shell - so shell metacharacters in jobid become executable commands. Per the CVSS vector the attack is network-reachable with low privileges required (PR:L) and no user interaction (UI:N), scoring 8.8, but it is strictly post-authentication and gated by a deliberately granted staff capability; no public exploit code has been identified at time of analysis and there is no CISA KEV listing. Vendor fixes are available in Koha 24.11.20, 25.05.15, 25.11.09, and 26.05.04.
Command injection in Ubiquiti's UniFi Protect AI Key lets a network-reachable, unauthenticated attacker (CVSS PR:N/UI:N) potentially execute arbitrary operating-system commands on the appliance, giving full loss of confidentiality, integrity, and availability. Per the supplied CVSS 3.1 vector (AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H, base 8.1), exploitation is high-complexity: the vendor conditions success on unspecified preconditions ('under certain conditions'), so attacks are not expected to succeed against every deployment and require the attacker to reach the device's network and satisfy environment-specific state. No public exploit code or confirmed active exploitation (CISA KEV) was identified at time of analysis, and no EPSS score was provided; the practical exposure is dominated by whether the AI Key is reachable from untrusted network segments.
OS command injection in EG Secure Solutions' SiteGuard Server Edition allows an attacker who already holds a low-privilege account or foothold on the host to execute arbitrary OS commands as root (Apache httpd/nginx on Linux or FreeBSD) or SYSTEM (IIS on Windows), achieving full compromise of the affected server. The CVSS 4.0 base score of 8.5 reflects a local attack vector (AV:L) with low privileges (PR:L), no user interaction (UI:N), low complexity (AC:L), and high confidentiality, integrity, and availability impact confined to the vulnerable system (VC:H/VI:H/VA:H, SC:N/SI:N/SA:N) - it is not remotely exploitable by an anonymous external client. No public exploit code and no confirmed active exploitation were identified at time of analysis; JPCERT/CC reported the issue and EG Secure Solutions and JVN have published advisories.
Command injection in the NetBird desktop client (versions 0.60.0 through 0.74.6) lets a malicious or compromised peer inject arbitrary OpenSSH directives - including command-executing ones - into the system-wide drop-in config of other clients' Linux, macOS, Windows and FreeBSD hosts. Exploitation requires the attacker to control an enrolled peer whose crafted FQDN or hostname containing a double quote is distributed to the victim, and the injected command runs only when a local user subsequently invokes ssh, so peer membership plus a local user action are both prerequisites rather than an unauthenticated internet-facing attack. NetBird 0.74.7 fixes the issue by rejecting peer names that are not valid DNS names; no public exploit identified at time of analysis, and the Android and iOS apps are not affected.
Unauthenticated OS command injection in the Collabora Online Built-in CODE Server Nextcloud app (richdocumentscode) before 26.4.303 allows remote attackers to execute arbitrary commands on the Nextcloud host via proxy.php, but only while the bundled CODE server process is not running - the window right after a restart, during startup, or following a crash. The CVSS 4.0 score is 9.2 and the CVE record attributes exploitation during September and October 2026, making this a genuine high-priority issue rather than a paper-tiger high-severity finding; the practical barrier is timing, not authentication, since the CVSS vectors carry PR:N/UI:N and an attacker can simply poll the endpoint until CODE is down. Standalone Collabora Online servers do not ship proxy.php and are explicitly not affected - the exposure is confined to Nextcloud instances with the richdocumentscode app, and the flaw is distinct from CVE-2025-66208.
OS command injection in the Samsung QA-Board backend allows unauthenticated remote attackers to execute arbitrary operating-system commands on the host running the service, yielding a CVSS 3.1 base score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) with full confidentiality, integrity and availability impact. The flaw, classified as CWE-78, affects all versions matched by the wildcard CPE cpe:2.3:a:samsung_open_source:qa-board:*:*:*:*:*:*:*:*, so no release line is known to be exempt. No public exploit code has been identified at time of analysis and the issue is not listed on CISA KEV, but the trivial remote/unauthenticated attack profile makes rapid patching the priority.
Command injection in the VNote note-taking application before version 4.3.0 lets an attacker-controlled note filename execute arbitrary commands as the victim user on Linux and macOS, but only when the victim opens that note and has configured an external program whose arguments are command-capable, such as vim/gvim's -c option. The root cause is SessionConfig::ExternalProgram::fetchCommand, which wraps the note path in double quotes and hands the whole string to QProcess::start(const QString&), so a double quote inside a POSIX filename terminates the quoted path and appends attacker-chosen arguments to the configured program. The vendor fixed the issue in 4.3.0; no public exploit identified at time of analysis, and the CVSS 3.1 vector (AV:L/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H, base 7.0) reflects full impact on confidentiality, integrity and availability once the narrow exploitation prerequisites are met.
Command injection in MultiTail 7.1.5 and earlier lets an attacker who can write content into a monitored log or stream execute arbitrary shell commands as the user running MultiTail, because matched text is concatenated into a command passed to sh -c with only partial escaping. Exploitation is gated by a deliberate, non-default configuration: the operator must have enabled the -ex/-eX execute-on-match feature with a filter regex, and the injected line must match that regex; because the CVSS vector is AV:L/PR:L/UI:R, this is a local, low-privileged, user-interaction-dependent issue with impact bounded by the privileges of the MultiTail process. Exploitation code is publicly available (CISA SSVC reports Exploitation: poc), but there is no evidence of confirmed active exploitation, and the assessed risk is moderate and internally consistent rather than alarming.
Command injection in MobSF v4.5.2 allows an attacker with access to the instance-and, for the /api/v1/scan path, a valid MobSF REST API key-to execute arbitrary shell commands via the /download_data/ endpoint, the /api/v1/scan REST API, or the Dynamic Analyzer ADB integration module (tests_common.py). The independent assessment scores this CVSS 8.8 (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H) rather than the 9.8 in the input because API-key authentication is enforced and the ADB path requires an active Android emulator or device. No public exploit identified at time of analysis, no CISA KEV listing, and EPSS is low at 0.71% (52nd percentile); MobSF is typically deployed internally, which reduces remote unauthenticated exposure.
OS command injection in Adobe Experience Manager 6.5 (through 6.5.25) and AEM 6.5 LTS (through SP3) allows an authenticated, low-privileged attacker to run arbitrary operating-system commands in the context of the current user, with changed scope (S:C) meaning the injected command executes outside the application's own security authority on the host. Rated CVSS 9.9 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H), the exploitation bar is low attack complexity with no user interaction; the only meaningful constraint is the need for a valid low-privilege AEM session, which is a very low bar in multi-user AEM author environments where content authors and other non-admin accounts are routinely issued. No public exploit code has been identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV); no EPSS score was supplied with the source data. This is a genuine high-priority vulnerability rather than a CVSS-inflated paper risk, and the specific vulnerable feature or injectable parameter is not named in the advisory summary, so patch validation and hunting should be prioritized on affected instances.
Command injection in Microsoft OneNote allows an unauthenticated remote attacker to execute code when a victim opens a specially crafted OneNote document. The issue is rated CVSS 8.8 (AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H) and affects Microsoft 365 Apps for Enterprise 16.0.1 per EUVD, with user interaction being the primary exploitation barrier. No public exploit was identified at time of analysis, and Microsoft has published an advisory and update path via the MSRC update guide.
Command injection in the Windows Shell lets remote, unauthenticated attackers bypass a Windows security feature and then execute arbitrary commands, yielding high impact to confidentiality, integrity, and availability. The attack is network-reachable with low complexity but requires victim interaction (CVSS UI:R) - the target must open or interact with attacker-supplied content such as a crafted file, shortcut, or link - which is why SSVC rates it non-automatable while still flagging total technical impact. The flaw affects Windows 10, Windows 11, and Windows Server 2012 R2 through 2025, plus Outlook.com, and is confirmed actively exploited (CISA KEV); given the KEV listing and the assessed high-end impact, this is a genuine top-tier patching priority rather than a high-CVSS-but-low-risk case.
Local privilege escalation in Windows Update enables an attacker who already holds a low-privileged local session to elevate to higher privileges by injecting special command elements into data processed by the Windows Update servicing component (CWE-77 command injection). This affects multiple Windows 11 releases (23H2, 24H2, 25H2, 26H1, 26H2) and Windows Server 2025, including Server Core, up to the fixed builds enumerated in Microsoft’s advisory, with a CVSS 3.1 score of 7.8 (AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H). Exploitation requires an existing local foothold and valid low-privileged credentials, and no public exploit has been identified at time of analysis; vendor patches are available.
Command injection in Visual Studio Code 1.0.0 through 1.141.0 allows an unauthenticated remote attacker to execute arbitrary code in the context of the editor when a developer opens attacker-controlled content such as a malicious workspace, folder, repository, or project configuration file. Exploitation depends on user interaction (a victim must open the crafted project), but is otherwise low-complexity with high confidentiality, integrity, and availability impact; no public exploit code has been identified at the time of analysis. Microsoft has released a vendor fix in Visual Studio Code 1.141.1.
OS command injection in KDE Dolphin 24.08.0 through 26.08.2 allows an attacker who can place a specially named directory on the victim's filesystem to run arbitrary commands with the victim's privileges when that folder's empty view background is double-clicked. The flaw lies in Dolphin substituting the folder path into the {path} placeholder of a shell command without neutralizing embedded double quotes and shell metacharacters, so a directory name such as one containing a quote plus `;`, `$()`, or backticks escapes the quoting in slotDoubleClickViewBackground. This is a genuine but interaction-gated local issue rather than a mass-exploitable remote flaw: exploitation requires the crafted directory to already exist (delivered via archive, network/share, USB, or cloned repository), an affected build running, and mandatory user interaction (the background double-click); no public exploit code was identified at time of analysis, and impact stays within the victim's user scope with no scope change.
OS command injection in rovr (a Python-based terminal file manager) through 0.10.3 on Windows allows a file name containing Windows shell metacharacters such as '&' to be passed unquoted to cmd.exe, executing arbitrary commands with the privileges of the user who opens that file via the open_file routine. Only Windows installations of rovr at or below 0.10.3 are affected; the vector is local and successful exploitation requires the victim to actively select and open an attacker-planted file, with the injected command running at the invoking user's existing privilege level (no elevation, no scope change). There is no CISA KEV listing and no public exploit identified at time of analysis; the CVSS 4.0 base score of 8.4 (High) reflects high confidentiality, integrity and availability impact, tempered by the local attack vector and mandatory user interaction.
OS command injection in Azure CycleCloud allows an authenticated local user with low privileges to execute arbitrary code on the affected host, but only if a victim performs a required user interaction. The CVSS 3.1 vector is AV:L/AC:L/PR:L/UI:R/S:U/C:H/I:H/A:H with a base score of 7.3, and the independent assessment rates this as a moderate-priority issue rather than a top-tier emergency. The available data provides no public exploit identified at time of analysis and no CISA KEV confirmation; Microsoft's MSRC update-guide entry is the provided remediation reference.
Authenticated OS command injection in the odt2spip plugin for SPIP before 4.4.0 allows author-level users to execute arbitrary commands as the web-server user by uploading a document with a crafted non-.odt filename through the document import form. The flaw survives sanitization in odt2spip_deplacer_fichier_upload() and LibreOffice::createCommand(), then triggers during LibreOffice conversion; exploitation requires a working LibreOffice backend and an author account. No public exploit was identified at time of analysis.
Authenticated users with workflow create/edit rights on self-hosted Heym deployments prior to 0.0.40 can escape the application's blacklist-based Python sandbox and execute arbitrary commands and file operations as the backend service account. The flaw lives in backend/app/services/python_tool_executor.py, where the allowlisted standard-library modules still expose OS, process, and filesystem primitives through public attributes, so tool source that passes _safe_import() and _validate_code_safety() is ultimately run through exec() with full application privileges (CVSS 8.8, CWE-78). Exploitation requires a low-privilege authenticated account (PR:L) with permission to create or edit a workflow containing an Agent custom Python tool; enabling self-registration makes that account trivially obtainable, though merely triggering a workflow from the portal does not grant code control. There is no public exploit identified at time of analysis and no confirmed active exploitation (not in CISA KEV), but the fix in 0.0.40 is available and the assessed risk is high for any multi-user deployment.
Unauthenticated remote command execution in Commvault CVFS, the file system component shipped with Commvault Grid (HyperScale X) deployments and exposed through Commvault Cloud, allows a network-reachable attacker to execute commands on the underlying system with high impact to confidentiality, integrity, and availability - Commvault scores it 9.3 under CVSS 4.0. Per the vendor CVSS vector (AV:N/AC:L/AT:N/PR:N/UI:N) the issue is remotely triggerable without credentials or user interaction, which makes any internet- or broadly reachable CVFS endpoint a priority exposure; the advisory does not name the specific feature, protocol endpoint, or configuration path that must be reached. No public exploit code and no confirmed active exploitation have been identified at time of analysis, and a fixed maintenance release is available from Commvault.
OS command injection in Wavlink WL-WN592AX6 firmware 250422 allows unauthenticated network attackers to execute arbitrary shell commands through the wmapd daemon's system function when they can reach the device management surface. The flaw is tracked as CVE-2026-109110 (CWE-78) with a 9.3 CVSS v4.0 score, and publicly available exploit code exists, though no confirmed active exploitation is indicated in the supplied data. Exploitation requires network access to the affected management interface; because this consumer range extender is typically LAN-facing, Internet exposure depends on WAN exposure or an attacker on the local/adjacent network.
Command injection in the Wavlink WL-WN530BE1-A wireless router (firmware 251117) allows an authenticated, low-privileged user of the web management interface to execute arbitrary shell commands on the device via the 'Type' parameter of /cgi-bin/uploadVpn.cgi, which is passed unsanitized into a popen() call. Because the flaw requires only a valid session (CVSS PR:L) with no user interaction (UI:N) and generally only LAN-side reachability to the management interface, the practical barrier is low: SOHO devices frequently ship with default or weak admin credentials. Publicly available exploit code exists (a published GitHub gist), but no confirmed active exploitation (CISA KEV) has been reported, and no vendor-released patch version is identified in the available data.
Command injection in the backup function of Teltonika Networks RUTOS and TSWOS allows an attacker who already holds high-privilege administrative access on an adjacent network segment to execute arbitrary OS commands on the device. Affected builds are RUTOS 7.22 before 7.24.6 and 7.25 before 7.25.2 on RUT, RUTX, RUTM, RUTC, ATRM, CAP, OXA, DAP, TRB, OTD and EBG devices, RUTOS 7.22 before 7.25.2 on TAP devices, and TSWOS 1.9 before 1.10.4 on TSW and SWM devices. No public exploit has been identified at time of analysis, and the assessed prerequisites (PR:H for administrator credentials plus AV:A for management-plane adjacency) mean exploitation is not reachable by an arbitrary internet attacker; patched firmware is available.
Remote OS command execution in IBM DataStage on Cloud Pak for Data 5.4.0.0 allows an authenticated attacker holding only low-privileged credentials to run arbitrary operating system commands on the underlying DataStage host, with high impact to confidentiality, integrity, and availability (CVSS 8.8, AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H). The flaw is an OS command injection (CWE-78) in which specially crafted input is passed to a system command without proper neutralization, so no user interaction is required and a single crafted request over the network is sufficient once valid access is obtained. Exploitation is gated by the need for existing credentials on the instance, and there is no public exploit code identified at time of analysis; IBM has published a fix via support node 7288649.
Authenticated remote code execution in IBM DataStage on Cloud Pak for Data 5.4.0.0 allows an attacker who already holds valid low-privilege credentials (CVSS PR:L, AV:N, UI:N) to run arbitrary operating system commands on the underlying host, resulting in high confidentiality, integrity, and availability impact (CVSS 8.8). The root cause is improper neutralization of special elements used in an OS command (CWE-94), meaning user-supplied input reaches a shell command without adequate sanitization. No public exploit code has been identified at time of analysis and the flaw is not confirmed actively exploited (not in CISA KEV), but a vendor patch is available per IBM's advisory, so unpatched 5.4.0.0 deployments exposed to multi-tenant users or broadly provisioned accounts should be prioritized for upgrade.
Command injection in the IBM Maximo Application Suite 'Location Service for Esri' component (versions 9.0, 9.1 and 9.2) allows a highly privileged attacker to execute arbitrary shell commands through a Kubernetes operator task, compromising confidentiality, integrity and availability beyond the component's own security boundary (CVSS 8.7). Exploitation is access-gated rather than trivial: the optional Esri location component must be deployed and enabled, and the attacker must already hold high privileges in the MAS environment (PR:H), but once those two conditions hold there is no user-interaction requirement and attack complexity is low, so injection is straightforward. No public exploit code and no confirmed active exploitation were identified at time of analysis.
Unauthenticated command injection in Edimax BR-6208AC firmware version 1.32 allows a remote attacker to execute arbitrary operating-system commands on the router by supplying crafted input through the readE2P argument of the /goform/formWlanMP endpoint. The issue is rated critical (CVSS 4.0 base 9.3; an independent assessment reaches the equivalent CVSS 3.1 vector AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) and publicly available exploit code exists, although there is no confirmation of active exploitation (device is not in CISA KEV). The vendor was contacted early and did not respond, and the product is no longer supported, so the decisive practical concern is permanent exposure: any BR-6208AC 1.32 whose web management interface is reachable over the WAN or from an untrusted network segment remains exploitable indefinitely.
Unauthenticated command injection in Edimax BR-6208AC firmware version 1.32 lets remote attackers execute arbitrary shell commands on the router by submitting crafted input to the tftpsIP parameter of the /goform/formAutoFWupgrade handler, which normally accepts a TFTP server address for firmware fetch operations. The vendor was contacted about the disclosure but did not respond, and the device is an end-of-life product no longer supported by the maintainer, so no vendor-released patch is expected; publicly available exploit code exists (a GitHub proof-of-concept under an IoT vulnerability repository) and the CVSS 4.0 score of 9.3 reflects a full confidentiality, integrity, and availability impact. The vulnerability is not listed in CISA KEV (no confirmed active exploitation reported), but pre-auth reachability of the HTTP management interface - particularly if WAN administration is enabled - makes this a high-priority exposure for anyone still running the device.
Authenticated content editors on Grav CMS sites running the Git Sync plugin versions 1.0.2 through 3.4.12 can execute arbitrary OS commands as the web-server account when the plugin's gravuser or gravfull author mode is enabled, by planting shell substitution payloads in their username, full name, or email profile fields that are later interpolated into a git commit --author call and reach PHP exec(). The vulnerability is rated 8.7 (CVSS 4.0) and maps to CWE-78; the assessed vector (CVSS 3.1 AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H) confirms that exploitation requires authenticated, low-privileged editor access but no victim interaction, with the additional gating condition that author mode must be explicitly turned on and must not be left at its default posture. No public exploit code has been identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV), so it is a credible, prioritizable bug rather than a drop-everything emergency - though any site with untrusted or broadly provisioned editor accounts and author mode enabled should treat it as an immediate command-execution risk.
Authenticated OS command injection in the Linksys E1700 router's embedded web management interface (firmware 1.0.0.4.003) lets a remote attacker with low-privilege credentials execute arbitrary operating-system commands on the device by manipulating the _6rd_prefix and _6rd_prefix_len parameters handled by setLinksysIpv6 at /goform/setLinksysIpv6. The flaw scores 8.6 under CVSS 4.0 (AV:N/AC:L/AT:N/PR:L/UI:N, E:P) and 8.8 under CVSS 3.1 (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), reflecting full confidentiality, integrity and availability impact on the router. Publicly available exploit code exists - including a published proof-of-concept write-up referenced from the CVE record - but there is no confirmed actively exploited (CISA KEV) status, and no vendor patch has been released because Linksys did not respond to the disclosure; practical exploitation requires reachability of the HTTP management interface plus valid (or default/weak) credentials, which on default deployments is typically LAN-side.
OS command injection in the Linksys E1700 router running firmware 1.0.0.4.003 lets an attacker execute arbitrary shell commands on the device by sending a crafted request to the /goform/setDMZ endpoint with a manipulated MAC_addr parameter. Exploitation requires network reachability to the web management interface (LAN-side by default, or internet-facing if remote administration is enabled) plus at least low-privilege authenticated access, and needs no user interaction; the vendor did not respond to the disclosure, so no patch is available. Publicly available exploit code exists (a proof-of-concept is published on GitHub), which raises the likelihood of opportunistic attacks against exposed devices.
OS command injection in Linksys E1700 firmware 1.0.0.4.003 allows authenticated low-privilege attackers to execute arbitrary OS commands on the device via crafted input to the static route parameters in /goform/addStaticRoute. Publicly available exploit code exists, increasing the risk of exploitation; no vendor patch has been released as Linksys did not respond to the disclosure. The vulnerability requires network access to the management interface and valid credentials, but successful exploitation grants full control of the router.
OS command injection in the Ministry of Justice General Directorate of Information Technology's UYAP Document Editor before version 5.4.20 allows an attacker to execute arbitrary operating-system commands on a victim's machine when a crafted document is opened in the application. The issue is local in nature (AV:L) and requires user interaction (UI:R) - the victim must open a malicious file - but no authentication or privileges are needed (PR:N), and successful exploitation yields high confidentiality, integrity, and availability impact, giving a base score of 7.8 (HIGH) on the CVSS 3.1 vector AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H. The flaw was reported by TR-CERT; no public exploit code has been identified and the issue is not listed in CISA KEV, so the practical risk is driven by phishing-style delivery of weaponized-looking documents rather than by remote scanning or automated mass exploitation.
OS command injection in mac-shell-mcp before 2.0.0 lets a connected MCP client - including a legitimate client driven by a malicious or prompt-injected LLM - escape the tool's command whitelist by appending shell metacharacters to the arguments of an otherwise permitted command such as pwd, causing forbidden commands to run as the server user without approval. Exploitation requires an existing MCP client session to the local server (CVSS 4.0 vector AV:L/PR:L/UI:N, mirroring the independent CVSS 3.1 assessment AV:L/PR:L/UI:N with high confidentiality, integrity and availability impact), so this is an authenticated, locally-scoped flaw rather than an internet-facing unauthenticated one, and impact is bounded by the privileges of the account the server runs under. No public exploit code has been identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV), but a vendor fix and upstream patch commit are available; EPSS data was not provided for this CVE.
OS command injection in Tuleap Enterprise Edition versions 17.0 through 17.5 allows an authenticated, low-privilege user to execute arbitrary commands on the underlying server, delivering a full compromise of confidentiality, integrity, and availability (CVSS 3.1 base 8.8, AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H). Because the assessed vector specifies PR:L, exploitation requires a valid account on the target instance - a registered project or site user is sufficient - and no user interaction is needed once that access is obtained, so unauthenticated attackers cannot trigger it directly. No public exploit code was identified at time of analysis and the CVE is not listed in CISA KEV; no EPSS score was supplied in the source data. Practical risk is therefore concentrated on internet-reachable Tuleap Enterprise Edition instances in the 17.0-17.5 range that allow self-registration or broad account provisioning, since such deployments effectively lower the credential barrier to exploitation.
Unauthenticated OS command injection in the Web Management Backend of TRENDnet TEW-751DR, TEW-752DRU, and TEW733GR routers (firmware 1.03B01/1.03B03) allows remote attackers to execute arbitrary shell commands on the device by sending crafted input in the 'service' argument handled by the fwrite function of pigwidgeon.cgi. Per the CVSS vector (PR:N/UI:N/AC:L) no authentication, user interaction, or non-default configuration is required; the principal limiting factor is reachability of the management interface, which is LAN-side by default and only exposed over WAN where remote administration is enabled or the router is internet-facing. Publicly available exploit code exists (a GitHub Proof-of-Concept targeting these models), so exposed units should be treated as compromisable with minimal attacker effort; EPSS and CISA KEV data were not provided, so confirmed active exploitation cannot be asserted. Vendor has confirmed the affected models reached end-of-life in 2011 and will not be remediated by a firmware fix.
OS command injection in GNU Emacs's TRAMP remote-file library (Emacs releases before 31.2, TRAMP through 2.8.2) lets an attacker who can get a crafted filename in front of a user execute arbitrary shell commands with that user's privileges, because tramp-user-regexp relies on an incomplete denylist of dangerous shell characters. This is a follow-on defect to CVE-2026-79992, whose fix did not cover all dangerous inputs. Exploitation is local, requires user interaction (the victim must open, complete, or otherwise have Emacs/TRAMP evaluate the malicious path), and only affects installations that use TRAMP remote-file functionality; no public exploit code and no confirmed active exploitation were identified at time of analysis.
Databasement instances before 1.8.2 allow an authenticated user to achieve remote code execution by injecting path-redirecting client options (such as mariadb-dump --result-file, -r, --tab, pg_dump --file, mongodump --out, or redis-cli --rdb/--eval) into the arguments passed to the backup dump binary. Shell quoting prevents classic OS command injection, but argument injection alone is sufficient: by directing the dump output to a web-served, PHP-executing path such as /app/public/index.php, the attacker writes code that the web server later executes. Authentication is required (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, base 8.8), publicly available exploit code exists, and success depends on the chosen output path being both web-accessible and interpretable as code; no evidence of confirmed active exploitation was provided.
OS command injection in pbi-cli 3.10.1 through 3.12.0 allows a local attacker to execute arbitrary commands with victim privileges when the victim opens a crafted Power BI .pbip project from a space-free filesystem path containing an '&' character. The flaw resides in desktop_sync.py, which passes unquoted paths to 'cmd /c start' during report write or reload, and exploitation requires user interaction (UI:P) with the crafted path. No public exploit code or active exploitation (CISA KEV) has been identified at time of analysis; CVSS 4.0 score is 7.3 (High), and an independent CVSS 3.1 assessment yields 7.8 (High).
OS command injection in the FILESystem:MOUNT:DRIve SCPI command handler of Tektronix 3 Series MDO oscilloscopes running firmware before 1.21.1 allows authenticated, low-privileged attackers who can reach the instrument's remote-control interface to execute arbitrary operating system commands with elevated privileges (CVSS 8.8, CWE-78, PR:L/UI:N). Any 3 Series MDO that exposes its SCPI control interface (LAN/VXI-11 or raw socket) to a lab or engineering network is in scope, and the attack requires no user interaction. No public exploit code and no CISA KEV listing were identified at time of analysis, and no EPSS score was supplied in the available data, so the realistic risk is concentrated on network-exposed instruments rather than mass exploitation.
OS command injection in ZoneMinder allows an authenticated user holding monitor-create or monitor-edit permission to execute arbitrary commands as the web server user by saving a Local-type monitor whose Device Path field contains shell metacharacters. The malicious stored Device value is later interpolated into shell commands in both Perl (qx()/backticks via zmdc.pl control paths and zmpkg.pl start/restart) and PHP (zmc daemon control helpers that relied on bypassable escapeshellcmd) when the monitor control or startup path runs. The issue is fixed in 1.39.2, 1.38.2, and 1.36.40, which reject any Device value not matching a strict /dev/ path pattern; no public exploit code has been identified at time of analysis, and exploitation is not confirmed actively exploited (KEV).
Remote command execution in ZoneMinder prior to 1.38.3 allows an authenticated user with monitor-create or monitor-edit rights to store a monitor Name containing shell metacharacters that break out of manual single-quote wrapping when a user with Events View permission exports that monitor's events, executing arbitrary commands as the web server user. Exploitation requires two authenticated roles and victim interaction (CVSS vector PR:L/UI:R), so this is not a default-reachable or mass-exploitable issue; no public exploit identified at time of analysis, and a vendor-released patch is available in 1.38.3.
Authenticated OS command injection in ZoneMinder versions prior to 1.38.3 allows a low-privilege user holding View Events or View Snapshots permission to run arbitrary operating system commands as the web server user. The flaw lives in exportEvents() in web/skins/classic/includes/export_functions.php, where the attacker-controlled exportFile request parameter is appended to a tar or zip command executed via exec() without shell escaping, so a crafted value can break out of the trailing directory argument. Exploitation requires a valid session (CVSS PR:L), a valid CSRF token for the export request, and the exportDetail option enabled; no public exploit code has been identified at time of analysis, and there is no confirmed active exploitation (CISA KEV), though the vendor's own GHSA advisory and fix commits confirm the issue is real and the EPSS-style risk is elevated by the low privilege threshold and high confidentiality, integrity, and availability impact (CVSS 8.8).
OS command injection in SparkyFitness versions prior to 1.6.0 allows an authenticated global administrator to execute arbitrary operating-system commands on the server by uploading a restore archive with a filename containing shell metacharacters. The vulnerability resides in the api/admin/backup/restore endpoint, which improperly interpolates the user-supplied filename into a shell command. Exploitation requires the highest-level application privilege (global admin), and no public exploit code has been identified at time of analysis. The issue is fixed in version 1.6.0.
Remote code execution in the Theme-one The Grid WordPress plugin through version 2.8.1 allows network attackers to inject code or commands via the plugin's grid-generation code path, with the supplied CVSS:3.1 vector indicating AV:N/AC:L/PR:N/UI:N (unauthenticated and no user interaction per the vector, though the exact endpoint and authentication prerequisite should be verified in the Patchstack advisory). The provided vector rates this as 10.0 Critical with full confidentiality, integrity, and availability impact, while our independent assessment adjusts scope from Changed to Unchanged but still considers impact High. No public exploit identified at time of analysis, and no KEV or EPSS data was provided.
Authenticated OS command injection in the WP Photo Album Plus WordPress plugin (all versions up to and including 9.3.02.002) lets any WordPress user granted the plugin-defined wppa_upload capability execute arbitrary operating-system commands as the web-server user, achieving full server-side code execution where ImageMagick is the active image backend. Rated CVSS 8.8 (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), the flaw is reachable through the update-photo AJAX action's value parameter, and no public exploit code was identified at time of analysis; the bug is not in CISA KEV. Because wppa_upload is a plugin-level privilege that administrators can delegate to lower-privileged roles, the effective attacker pool can extend well beyond site administrators on misconfigured sites.
Command injection (CWE-78) and stored XSS in Wheels CLI 4.0.0 through 4.1.1 allow an attacker who can influence inputs processed by the tool to run arbitrary shell commands as the SSH deploy user on deploy hosts and on the CI runner during builds. Deploy release/version/rollback values, package removal/destroy and extraction paths, and generator names are the primary injection surfaces, with unencoded scaffold 'show' views (since 4.1.0) and generated admin views permitting stored XSS; trigger inputs can come from CI variables, Git branch names, downloaded packages, or an AI agent using the CLI's MCP tools. The flaw is fixed in 4.1.2, and existing affected generated views also require a documented manual edit; no public exploit code has been identified at time of analysis.
Command injection in pfSense CE 2.7.0 through 2.9.0 allows authenticated users who hold only the Captive Portal HA privilege to execute commands as root by manipulating backwardsyncip to point at an attacker-controlled XMLRPC server and returning connected_users entries with shell metacharacters in ip or mac fields. Exploitation is privilege-gated: it requires the specific Captive Portal HA WebGUI permission, use of Captive Portal high-availability sync, and the ability to influence the synced connected_users data. Publicly available exploit code exists, but no CISA KEV confirmed active exploitation is identified; this is a credible privilege-gated priority rather than a mass-exploitation emergency.
Command injection in Netgate pfSense CE 2.4.5 through 2.9.0 lets an authenticated WebGUI user holding the Advanced Admin Access privilege execute arbitrary commands as root by placing newline or $(...) sequences in the Login Protection (sshguard) threshold, block time, or detection time fields, which are written unvalidated to sshguard.conf and sourced as root when the service restarts. The flaw is privilege-gated rather than mass-exploitable: the CVSS vector specifies PR:H, so only already-administrative accounts can trigger it, with no user interaction required beyond the attacker's own session. Publicly available exploit code exists (referenced in the CVE record), and the root cause is CWE-78 OS command injection.
Authenticated VPN users can execute arbitrary commands as root on pfSense CE through 2.9.0 by exploiting an OS command injection in the OpenVPN connect/disconnect script, openvpn.connect_async.sh. The attack requires a valid VPN account authenticated via RADIUS or LDAP and a username containing single quotes and shell metacharacters that break out of an eval'd pfctl anchor name. Publicly available exploit code exists (reported by VulnCheck), though no active exploitation has been confirmed by CISA KEV. This is a credible high-priority issue for any pfSense CE deployment that uses OpenVPN with external authentication.
OS command injection in Netgate pfSense CE 2.7.0 through 2.9.0 lets an authenticated web GUI user who holds the Traffic Shaper Limiters privilege execute arbitrary commands as root, turning a narrowly delegated configuration permission into full appliance compromise. The flaw is triggered when an operator edits a limiter and injects a crafted hidden 'number' value and then deletes that limiter, causing delete_queue() to pass the unquoted, attacker-controlled value to dnctl via /bin/sh running as root. Publicly available exploit code exists (VulnCheck advisory plus Abraxas Labs research and a PoC repository), and while the issue is not listed in CISA KEV as actively exploited, exploitation is gated on prior authenticated access with the Limiters privilege (PR:H) and on the attacker completing the limiter edit-then-delete workflow.
Command injection in Netgate pfSense CE through 2.9.0 allows an authenticated user with Virtual IP (CARP) edit privileges to execute commands as root by storing a malicious VHID value that is later interpolated into ifconfig calls or CARP status rendering. Publicly available exploit code exists, and the independent assessment rates this as a privilege-gated high-impact issue rather than an unauthenticated remote compromise; CISA KEV does not confirm active exploitation in the supplied data. Affected deployments are those where administrators or delegated operators can create or modify CARP VIPs, narrowing realistic risk to insiders or attackers who already hold privileged firewall UI access.
OS command injection in Netgate pfSense CE (through 2.9.0) allows remote attackers holding an authenticated web-GUI account with gateway edit privileges to execute arbitrary commands as root by planting shell metacharacters in the zone index of an IPv6 link-local gateway Monitor IP, which start_dpinger() passes unescaped into the dpinger shell command when gateway monitoring is applied. Exploitation requires high privileges (CVSS 4.0 PR:H) and no victim interaction (UI:N), but yields high confidentiality, integrity and availability impact on the firewall itself; publicly available exploit code exists (abraxas/pfsense-gw-ipv6-zone) with no confirmed active exploitation reported. This is a genuine but access-gated priority: unauthenticated or low-privileged users cannot reach the vulnerable gateway-edit path, so exposure is limited to deployments where such configuration rights are delegated.
Command injection in Netgate pfSense CE through 2.9.0 allows remote authenticated users with limited 'Wireless Edit' or 'Interface Assignments' GUI privileges to execute arbitrary commands as root. The flaw is triggered by supplying a crafted wireless parent interface name such as ath0;<cmd> through interfaces_wireless_edit.php or interfaces_assign.php, which interface_wireless_clone() interpolates unescaped into an ifconfig shell command. Publicly available exploit code exists, and the assessed impact is high for confidentiality, integrity, and availability, though exploitation requires valid credentials with one of those specific privileges and access to the management interface.
OS command injection in Netgate pfSense CE 2.7.0 through 2.9.0 allows an authenticated web-interface user who holds the QinQ edit privilege to execute arbitrary commands as root. The attacker saves a crafted parent interface ('if') value and then re-edits the QinQ entry, causing that unvalidated value to be interpolated into an ifconfig shell command; publicly available exploit code exists, and no confirmed active exploitation (CISA KEV) has been reported. The principal constraint is the high privilege requirement - only operators already trusted with interface configuration can reach the vulnerable code path - and risk is further reduced where the management UI is not exposed to untrusted networks or lower-privileged users.
OS command injection in Unitronics UniLogic (versions prior to 1.41.329) and the UniStream Pro B10/C10 controllers it programs allows an authenticated low-privilege attacker who can reach a vulnerable input field to execute arbitrary operating-system commands on the affected device, resulting in full compromise of confidentiality and integrity plus partial availability impact. The flaw is network-reachable with low attack complexity and no user interaction, but requires valid credentials (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), which limits practical exposure to devices whose programming or web interface is reachable from an untrusted network segment. No public exploit code or confirmed active exploitation was identified at time of analysis; the issue was reported by Israel's National Cyber Directorate (INCD).
OS command injection in Wavlink WL-WN530BE2-A firmware up to M30BE2_V250604 lets remote attackers reach the api.cgi login handler (api=info&client=app, function sub_12580) and inject shell metacharacters through the value the firmware treats as the client address (REMOTE_ADDR), potentially yielding arbitrary command execution with root-level privileges on the device. The vendor has confirmed the finding and shipped a fix in firmware 2026-09-24-841a51c6d5_tclinux.bin, which replaces the system() call with execvp and length-limits/null-checks the client address; publicly available exploit code exists (a GitHub PoC), though the weakness is not confirmed as actively exploited (CISA KEV). Exploitation is not as trivial as the raw CVSS 4.0 score of 9.3 implies: REMOTE_ADDR is conventionally derived from the TCP source address, so a real-world attacker likely needs the web server to trust a client-supplied address header or needs an on-path/proxy position, which materially narrows the population of reachable targets despite PR:N in the published vector.
Unauthenticated remote command execution in QNAP HybridDesk Station (HD Station) lets network attackers run arbitrary operating system commands on affected QNAP NAS devices that have the optional desktop/multimedia add-on installed. The root cause is an OS command injection (CWE-78) rated 9.3 under CVSS v4.0 (AV:N/AC:L/PR:N/UI:N), so no credentials, user interaction, or special preconditions are required; the vendor has fixed it in HybridDesk Station 4.2.23 (4.2 branch) and 4.3.4 (4.3 branch). No public exploit code or confirmed active exploitation has been identified at time of analysis, but real-world risk is governed by whether HD Station is installed and whether its interface is reachable from the attacker's network position (LAN versus internet-facing), as the specific vulnerable feature and input parameter have not been disclosed.
OS command injection in the pfSense-pkg-frr package (2.0.2_1 and source revision 2.1.2_1) lets an authenticated WebGUI user holding only the low-privilege page-status-frr role execute arbitrary commands as root on the firewall. The flaw sits in the FRR status filter on status_frr.php, where a crafted filter value escapes a single-quoted sed expression that concatenates escapeshellarg() output before being evaluated through a popen() pipeline. Exploitation requires the optional FRR package to be installed plus a valid authenticated session with the page-status-frr privilege (CVSS 3.1: AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, score 8.8), so it is a post-authentication privilege escalation rather than a pre-auth compromise; there is no public exploit identified at time of analysis, and no CISA KEV listing. Updated FRR packages resolve the issue for pfSense Plus 26.07/26.03.1 and CE 2.9.0/2.8.1.
Arbitrary OS command execution on hosts running MoviePilot through 3.1.5 is achievable when the product's LLM agent ingests attacker-controlled content via browse_webpage, search_web, or torrent metadata parsing and is induced to call the unrestricted ExecuteCommandTool, which runs shell commands through asyncio.create_subprocess_shell under the admin session's is_admin context. The flaw is a prompt-injection-to-RCE chain (CWE-1427) that bypasses the validate_command_safety denylist by setting confirm_dangerous=true; the attacker needs no MoviePilot credentials (CVSS PR:N), but an administrator must interact with the poisoned content (UI:P in CVSS:4.0, UI:R in the assessed CVSS:3.1) and the model must be successfully coerced, so the issue is conditionally gated rather than a mass-exploitation emergency. Severity is 8.7 in CVSS:4.0 and 8.7 in the independently assessed CVSS:3.1; no public exploit code and no CISA KEV entry are associated as of this analysis.
Authenticated attackers with high privileges on SonicWall Network Security Manager (NSM) on-prem can execute arbitrary OS commands via an OS command injection flaw (CWE-78), achieving remote code execution. The vulnerability affects NSM on-prem versions 4.3.1 and earlier, requires a privileged session (PR:H), and is assessed with high attack complexity (AC:H) due to unenumerated specific conditions. No public exploit code or active exploitation has been identified (CISA KEV status: none; SSVC exploitation: none), and the EPSS score is 0.54% (44th percentile), indicating a moderate-priority risk rather than an urgent emergency.
Command injection in the Markdown Preview Enhanced VS Code extension (which bundles the crossnote library) on Windows allows an attacker-controlled Markdown link to execute arbitrary commands with the privileges of the VS Code user. When a victim opens an untrusted Markdown file, previews it, and clicks a crafted external or unsupported-scheme link, the decoded link target is passed to a Windows file-opening helper that invokes child_process.exec with the start command, so embedded cmd.exe metacharacters are interpreted as command separators. The flaw affects versions prior to 0.8.28 (crossnote prior to 0.9.29), requires user interaction (opening the file and one link click), needs no workspace trust, no script execution, and no additional files, and earns a CVSS 3.1 base score of 8.8; no public exploit code was identified at time of analysis and the issue is not in CISA KEV.
Unauthenticated OS command injection in the D-Link DIR-868L router running firmware 2.01b05 lets remote attackers execute arbitrary shell commands on the device by sending a crafted SOAPACTION argument to the HNAP handler exposed through /htdocs/cgibin. The flaw is remotely reachable with no credentials, no user interaction, and no non-default configuration required (CVSS v4.0 base 9.3, assessed v3.1 as AV:N/AC:L/PR:N/UI:N/C:H/I:H/A:H), so any installation exposing the HNAP management interface is at risk; LAN-only deployments are largely confined to local-network exploitation, while devices with WAN-side remote management enabled are directly attackable from the internet. Publicly available exploit code exists (a detailed advisory and PoC are published on GitHub), which raises the practical urgency, though active exploitation is not confirmed by CISA KEV, and the product is an end-of-life model for which no vendor-released patch has been identified.
OS command injection in the phcd phone-home daemon of Juniper Networks Junos OS and Junos OS Evolved lets an adjacent, unauthenticated attacker who operates a rogue DHCP server execute arbitrary commands as root on affected EX Series, SRX Series, and QFX Series devices. Exploitation is not remotely routable: the attacker must share a Layer-2 broadcast domain with the target and the device must be actively running phone-home / zero-touch provisioning (ZTP), which is why an independent assessment rates the attack complexity higher than the vendor's 8.7 score despite the high confidentiality, integrity, and availability impact (PR:N/UI:N once those conditions hold). No public exploit code has been identified at time of analysis and the flaw is not confirmed actively exploited (no CISA KEV entry), though NVIDIA/ENISA EUVD tracking and a full Juniper advisory (JSA112187) with vendor patches indicate this is a genuinely high-priority, patchable issue.
Authenticated, low-privileged TACACS+ users on Juniper Networks Junos OS Evolved can escalate to root and execute arbitrary OS commands, but only when the periodic TACACS+ authorization-refresh feature is enabled - an OS command injection flaw (CWE-78) that requires both a valid low-privileged TACACS+ account and that specific configuration to be active. The vendor rates it 9.3, while our independent assessment scores it CVSS:3.1 9.0 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H), reflecting that exploitation is not possible by an anonymous remote attacker despite the vendor's PR:N vector. There is no public exploit identified at time of analysis, and the flaw does not trigger on devices using TACACS+ without authorization-refresh, or on devices using RADIUS or local authentication.
OS command injection in the Junos OS CLI lets an authenticated, low-privileged local user escalate to full root control of the device by passing unsanitized arguments to the 'request csds' operational command. Only Junos OS 24.4 (from 24.4R1 up to before 24.4R2-S4), 25.2 (before 25.2R2-S1), and 25.4 (before 25.4R1-S3) are affected; releases before 24.4R1 are explicitly unaffected, and the attack is strictly local with no network or unauthenticated path. There is no public exploit identified at time of analysis, and Juniper has released fixed versions (24.4R2-S4, 25.2R2-S1, 25.4R1-S3), making this a serious but access-gated issue rather than an internet-facing emergency.
Unauthenticated OS command injection in Juniper Networks Junos OS Evolved's bootzd zero-touch provisioning (ZTP) daemon allows adjacent attackers on the management LAN to execute arbitrary commands as root via a crafted DHCP response. The vulnerability has a CVSS 4.0 base score of 9.3 and requires the target device to be actively in ZTP mode; no public exploit code is identified at time of analysis.
Unauthenticated OS command injection in the bootzd zero-touch provisioning (ZTP) daemon of Juniper Networks Junos OS Evolved on PTX Series routers allows an attacker positioned on the management LAN to execute arbitrary commands as root by sending a crafted gRPC response to a device that is actively provisioning. The vulnerability carries a CVSS 3.1 base score of 8.8 (AV:A/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H), reflecting low attack complexity, no authentication or user interaction requirements, and full confidentiality, integrity, and availability impact; the adjacent-network attack vector (AV:A) means the device is not exploitable directly from the internet. No public exploit code has been identified at time of analysis, no CISA KEV listing or EPSS score was provided, and exposure is materially reduced where management traffic runs over a segmented or physically isolated out-of-band network and where devices are not in an active ZTP state.
mise (a polyglot dev tool manager for node, python, cmake, terraform, and others) before version 2026.7.14 lets a malicious local project inject interpreter arguments through an untrusted .mise.toml [settings] file, causing attacker-chosen commands to execute with the developer's privileges. In default non-safe mode, mise applies local project settings before trust approval, and the unix_default_file_shell_args, unix_default_inline_shell_args, windows_default_file_shell_args, and windows_default_inline_shell_args fields were missing the global_only marker, so they survived strip_local_only_settings. When a developer later runs a trusted task, postinstall directive, template exec, or hook that has no explicit shell, mise appends the legitimate command after the attacker-controlled interpreter arguments, and the attacker's command runs first despite the config never being trusted. Exploitation is local and requires user interaction (the victim must clone/open and run the affected task); there is no public exploit code and no confirmed active exploitation at time of analysis, and this is best characterized as a conditional local-escalation risk rather than a mass-exploitation threat. Fixed in 2026.7.14.
A never-ending loop condition in curl and libcurl 8.10.0 through 8.22.0 lets a malicious, compromised, or on-path-manipulated server hang the client by omitting an expected part of its response, producing availability-only impact (the affected transfer or process spins indefinitely) with no confidentiality or integrity effect. The curl project rates this flaw LOW severity, and it is one of twenty-four CVEs disclosed alongside the curl 8.23.0 release. Exploitation requires the victim to connect outbound to the attacker's endpoint, since curl is a client and cannot be attacked purely remotely against a listening service; the exploit is not confirmed as actively exploited (not in CISA KEV), no public exploit code has been identified at time of analysis, and no EPSS score was provided in the source data.
Stack exhaustion in libcurl's MIME handling can crash applications that construct deeply nested multipart structures: curl_mime_free() recurses once per nesting level, so a sufficiently deep curl_mime_subparts() hierarchy overflows the stack and aborts the process. Affected are libcurl/curl versions 7.56.0 through 8.22.0, fixed in 8.23.0; the curl project classifies this advisory as LOW severity, with the impact limited to denial of service (no code execution, no data disclosure). The assessed vector (CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L) reflects that exploitation is confined to libcurl-based applications that let untrusted input dictate MIME nesting depth, and there is no public exploit identified at time of analysis.
OS command injection in the thumbnail Python package through version 1.5 allows attackers who control a filename passed to generate_thumbnail() to execute arbitrary shell commands with the privileges of the Python process. The vulnerable code concatenates user-influenced values (input, output, width, height, quality, trim) directly into os.system() calls invoking ffmpeg, convert or unoconv, so shell metacharacters such as ';', backticks or $(...) in a filename break out of the intended command. The package is a library with no network listener of its own, so real-world reachability depends on the consuming application exposing untrusted filenames - classically a file-upload feature - and on ffmpeg/convert/unoconv being installed for the relevant code branch; where those conditions hold, no authentication or user interaction beyond submitting the malicious filename is required. An upstream fix exists in pull request #9, and no public exploit code has been identified at time of analysis.
OS command injection in Stills (shahkashani/stills), a Node.js media-processing library, before version 25.24.2 allows crafted media file names containing shell command-substitution syntax such as $(...) or backticks to be executed as operating system commands with the privileges of the Node.js process. The flaw is reachable whenever the library shells out to ffmpeg, ffprobe, gifsicle, or convert on an attacker-influenced path, and the CVSS 4.0 vector (9.3) describes an unauthenticated, low-complexity network attack (PR:N/UI:N) yielding high confidentiality, integrity, and availability impact; our independent assessment scores it CVSS:3.1 8.8/9.8-class (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) but cautions that real-world risk is materially lower in deployments where media file names are operator- or system-generated rather than user-supplied. No public exploit code or confirmed active exploitation was identified at time of analysis, and a vendor patch (25.24.2, via PRs #29/#30 and associated commits) is available.
OS command injection in @offactory/compress-fonts before 1.1.1 lets an attacker execute arbitrary shell commands with the privileges of the Node.js process by getting a font file with shell metacharacters in its filename into a directory that is being compressed. The supplied CVSS 4.0 vector indicates no authentication is required (PR:N) and high confidentiality, integrity, and availability impact to the vulnerable component, but the attack vector is local (AV:L) and requires a user or automated build to initiate the FontForge conversion (UI:P), so real-world impact is bounded by who can place files in the processed source directory and by the privileges under which the build runs. No public exploit code has been identified at time of analysis and the issue is not listed in CISA KEV; a vendor fix is available in version 1.1.1, which replaces execSync with execFileSync.
OS command injection in Turnpress through 0.4.2 allows arbitrary shell commands to be executed with the victim user's privileges when the tool's convert or create operation processes a .docx file whose filename contains shell metacharacters. The flaw sits in convertDocxToHtml (src/converter.ts), which invokes pandoc through execa with shell:true, so the crafted filename is interpreted by the shell rather than treated as a literal argument; exploitation is local (AV:L) and requires the victim to run a convert/create command on the malicious file (UI:R), and no privileges are needed to craft the filename. Impact is high for confidentiality, integrity, and availability on the host running the conversion, but it is limited to the invoking user's privileges with no scope change, and no public exploit code or confirmed in-the-wild exploitation has been identified at time of analysis; vendor fix is available via an upstream pull request that removes shell:true.
OS command injection on Windows in Beganovich Snappdf versions 1.1.0 through 6.0.x lets remote attackers execute arbitrary operating-system commands as the PHP process by feeding cmd.exe metacharacters (such as '&') into the URL passed to setUrl(). Because the vulnerable code path is built only when the library runs on Windows, the issue is exploitable only when the integrating PHP application runs on Windows and passes attacker-influenced data into setUrl(); a fully static or trusted URL is not exploitable. The CVSS v4.0 vector (AV:N/PR:N/UI:N) indicates the injection can be triggered remotely without authentication, giving high confidentiality, integrity, and availability impact on the host, although no public exploit code or confirmed in-the-wild exploitation has been identified at time of analysis.
OS command injection in the Video-stitch Node.js library (ArsalanDotMe) before 2.0.0-beta.1 allows unauthenticated remote attackers to run arbitrary shell commands with the privileges of the Node.js process by supplying file names containing shell metacharacters to the concat(), cut(), or merge() functions. The vulnerable code interpolated those values unquoted into shell-executed ffmpeg command lines (shelljs.exec) in lib/videoconcat.js and lib/videocut.js, so exposure is realized whenever an embedding application routes untrusted data into the output/clip name or fileName parameters; applications that only pass trusted, internally generated file names are not exploitable. CVSS 4.0 rates this 9.3 (Critical) with PR:N and UI:N, and no public exploit code or confirmed active exploitation was identified at time of analysis; a fix is available in version 2.0.0-beta.1.
OS command injection in nodejs-whisper through 0.3.1 lets remote, unauthenticated attackers execute arbitrary operating-system commands with the privileges of the Node.js process by supplying a crafted audio file path. The vulnerable convertToWavType() function builds an ffmpeg command as a string, interpolating the caller-provided path between double quotes before handing it to shelljs exec(); a filename containing $() command substitution or a closing double quote breaks out of the quoted argument and is interpreted by the shell. An upstream fix exists in pull request #259, but no tagged patched release has been independently confirmed, and no public exploit code was identified at time of analysis - real-world exposure is limited to applications that pass attacker-influenced, unsanitized filenames into the transcription/convert entry points while ffmpeg is installed on the host.
OS command injection in the transcribe2texts Node.js library through version 1.0.6 lets unauthenticated remote attackers execute arbitrary shell commands with the privileges of the Node.js process by supplying crafted audioPath values to transcribeAudio(). The flaw is a library sink rather than a directly exposed service: it is exploitable only where an integrating application passes attacker-influenced data into audioPath, which is concatenated unescaped into a python3 command line and run through child_process.exec(); applications that hardcode or strictly validate that path are not exposed. The critical 9.3 (CVSS 4.0) severity reflects network-reachable, low-complexity, no-privilege, no-interaction exploitation with high confidentiality, integrity and availability impact to the host, and while an upstream fix exists, no public exploit code has been identified at time of analysis.
OS command injection in xhtml2pdf before 0.2.19 lets an attacker execute arbitrary OS commands as the invoking user when the pisa command-line tool is run with the -s/--start-viewer option against a file whose name contains a double quote plus shell metacharacters. The flaw lives in pisa.startViewer(), which interpolates an unescaped filename into os.system('open "%s"'); because the vulnerable path is only reached in viewer-launch mode, the default programmatic HTML-to-PDF conversion API is not affected. The vector is local and requires a user (or job) to actually launch the viewer, and no public exploit code has been identified at time of analysis - confirmed actively exploited status is not present in the available data.
OS command injection in Runnerty's @runnerty/executor-decompress package (all versions through 3.1.1) lets an attacker who can control the name of an archive file processed by a compress_file/decompress task execute arbitrary shell commands with the privileges of the Runnerty process. The package builds shell command strings from templates that interpolate the archive filename and then hands them to shelljs exec(), so a filename containing $() command substitution or unquoted shell metacharacters - for example one dropped into a watched or ingested directory - is interpreted by the shell instead of being treated as a literal path. Exploitation requires low-privileged access sufficient to name files in that directory (PR:L) and no user interaction; impact to confidentiality, integrity, and availability is high but is bounded to the Runnerty process's privilege level, and no public exploit code or confirmed active exploitation was identified at time of analysis.
OS command injection in the whisper-node npm package through version 1.1.1 allows remote, unauthenticated attackers to run arbitrary shell commands with the privileges of the embedding Node.js process whenever an application passes attacker-controlled data into whisper()'s filePath argument or the whisperOptions.language (and potentially modelPath) options. The vulnerable code path is the default in all versions up to and including 1.1.1, with no special configuration required, although real-world exposure is gated by how the library is used: whisper-node is a library, not a network service, so only applications that feed it untrusted input are remotely exploitable, and impact is bounded by the Node.js process privileges. No public exploit code has been identified at time of analysis, and there is no indication of confirmed active exploitation (CISA KEV); the upstream fix appears in GitHub PR #64 but a released patched version could not be independently confirmed from the available data.
OS command injection in @mborne/extract versions before 0.2.1 lets an attacker execute arbitrary shell commands with the Node.js process's privileges by supplying shell metacharacters in the archivePath or targetDir options passed to extract(). The flaw is unauthenticated (CVSS 4.0 vector AV:N/AC:L/AT:N/PR:N/UI:N) and scored 9.3, with full confidentiality, integrity, and availability impact to the host process, though it is a library with no standalone attack surface - it is only reachable when the integrating application forwards attacker-controlled values into one of the affected extractor paths (unzip, 7z, tar-bz2, or gz; the 7z path additionally requires the 7z binary). No public exploit code was identified at time of analysis, so exploitation is contingent on the host application's input handling rather than on any inherently exposed service.
OS command injection in doc-recog through 1.1.6 allows remote attackers to execute arbitrary commands with Node.js process privileges by supplying crafted document paths containing $(...) or backtick command substitution to the recog() or tag() functions, which flow to a shell-based pdftotext exec call in DocData._readPDF. The CVSS vector indicates no authentication (PR:N) and no user interaction (UI:N), but real-world exploitability hinges on whether a consuming application passes attacker-controlled paths into the library; if paths are hardcoded or validated downstream, the flaw is not reachable. No public exploit code has been identified at time of analysis, and the upstream fix (PR #2) switches to execFile with an argument array to eliminate shell interpretation.
OS command injection in buildebooks (wastemobile) versions before 0.2.2 lets a crafted EPUB file name or project folder name break out of the shell command string built in lib/kindle.js and execute arbitrary operating-system commands with the privileges of the user running the build. Exploitation is local rather than network-reachable: the victim must process an attacker-supplied book project with the Kindle output path enabled (kindle: true), which satisfies the user-interaction prerequisite, and no privileges are needed on the attacker's side per the assessed vector (PR:N). No public exploit code or confirmed in-the-wild activity has been identified at time of analysis, but the affected code path is straightforward (a naive string concatenation into child_process.exec) and a vendor fix is available in 0.2.2.
OS command injection in the Python library pypdftk (through version 0.5) lets attackers who control the pdf_path or out_file arguments of fill_form(), dump_data_fields(), or pdftk_cmd_util() execute arbitrary operating-system commands with the privileges of the hosting Python process. The three functions concatenate caller-supplied paths into a command string using '%s' formatting and pass it to subprocess.Popen with shell=True, so shell metacharacters embedded in a filename or path are interpreted by the shell rather than treated as literal arguments. CVSS v4.0 scores the flaw 9.3 (AV:N/AC:L/AT:N/PR:N/UI:N), and the independent assessment concurs with CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H, but real-world reachability is bounded by the calling application - pypdftk is a library, not a network service, so exploitation requires the embedding application to forward untrusted input into those parameters, and no public exploit code has been identified at time of analysis. No CISA KEV entry exists for this CVE, and a source-level fix is available from upstream via pull request 57, which replaces string command construction with argument lists.
OS command injection in webp-converter before 3.0.0 lets unauthenticated remote attackers execute arbitrary operating system commands as the Node.js process when a host application routes attacker-controlled file names or option strings into the library’s cwebp, dwebp, gwebp, or webpmux helper functions. The flaw carries a CVSS 4.0 score of 9.3 (with an independent CVSS 3.1 assessment of 9.8), but because webp-converter is a library rather than a listening service, real-world risk depends on whether the integrating application actually passes untrusted input into the vulnerable code path; applications that use only server-controlled, fixed paths and options are not exposed. No public exploit code has been identified at the time of analysis, and the vendor has released version 3.0.0 to fix the issue.
OS command injection in the npm package git-diff (danday74/git-diff) before 2.0.7 lets unauthenticated attackers execute arbitrary shell commands with the privileges of the Node.js process by passing crafted document strings into the gitDiff API. The library hashed each input with a shell command of the form `printf "..." | git hash-object -w --stdin` evaluated through shelljs.exec, so payloads containing $(cmd) or backtick substitutions survived the JSON.stringify escaping and were executed. The vendor has published an upstream fix that removes the shell invocation entirely, and no public exploit code has been identified at time of analysis; the assessed CVSS 4.0 base score is 9.3 (Critical) for maximal confidentiality, integrity and availability impact to the vulnerable system.
Unauthenticated OS command injection in TR7 Cyber Defense Inc.'s TR7 ASP appliance (both virtual and physical deployments) allows any attacker who can reach the appliance's exposed service to execute operating-system commands without supplying credentials or user interaction. The flaw is rooted in a missing authentication check for a critical function (CWE-306) and carries a CVSS 3.1 base score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H), meaning full loss of confidentiality, integrity, and availability of the appliance if exploited. All TR7 ASP installations running 1.0.40.236 up to (but not including) 1.0.1554.2331, or 1.6.42.239 up to (but not including) 1.6.1554.2331, are affected; exposure is the primary gating factor, since appliances segmented onto trusted management VLANs are far harder to reach. No public exploit code and no CISA KEV entry have been identified at time of analysis, and no EPSS score was provided with the source data.
Remote unauthenticated OS command injection in TR7 ASP allows an attacker who can reach the appliance over the network to execute arbitrary operating system commands with the privileges of the vulnerable service, yielding full loss of confidentiality, integrity, and availability. Affected builds are TR7 ASP from 1.0.40.236 before 1.0.1554.2331 and from 1.6.42.239 before 1.6.1554.2331, covering both virtual-appliance and physical-appliance deployments. The flaw carries a CVSS 3.1 base score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) per the assessed vector, meaning no authentication, no user interaction, and low attack complexity; no public exploit code has been identified at time of analysis and the issue is not listed as confirmed actively exploited (CISA KEV). The primary limiting factors are version and network exposure: only the specified build ranges are affected and the appliance must be reachable by the attacker.
OS command injection in the Koha integrated library system lets an authenticated staff account holding the tools => schedule_tasks permission execute arbitrary operating-system commands as the Koha web/application user. The flaw lives in the tools/scheduler.pl handler: an op=cud-delete request passes the unvalidated jobid parameter into C4::Scheduler::remove_at_job, which hands it to Schedule::At::remove, where the value is spliced into an atrm command string and run through a single-argument system() call that invokes the shell - so shell metacharacters in jobid become executable commands. Per the CVSS vector the attack is network-reachable with low privileges required (PR:L) and no user interaction (UI:N), scoring 8.8, but it is strictly post-authentication and gated by a deliberately granted staff capability; no public exploit code has been identified at time of analysis and there is no CISA KEV listing. Vendor fixes are available in Koha 24.11.20, 25.05.15, 25.11.09, and 26.05.04.
Command injection in Ubiquiti's UniFi Protect AI Key lets a network-reachable, unauthenticated attacker (CVSS PR:N/UI:N) potentially execute arbitrary operating-system commands on the appliance, giving full loss of confidentiality, integrity, and availability. Per the supplied CVSS 3.1 vector (AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H, base 8.1), exploitation is high-complexity: the vendor conditions success on unspecified preconditions ('under certain conditions'), so attacks are not expected to succeed against every deployment and require the attacker to reach the device's network and satisfy environment-specific state. No public exploit code or confirmed active exploitation (CISA KEV) was identified at time of analysis, and no EPSS score was provided; the practical exposure is dominated by whether the AI Key is reachable from untrusted network segments.
OS command injection in EG Secure Solutions' SiteGuard Server Edition allows an attacker who already holds a low-privilege account or foothold on the host to execute arbitrary OS commands as root (Apache httpd/nginx on Linux or FreeBSD) or SYSTEM (IIS on Windows), achieving full compromise of the affected server. The CVSS 4.0 base score of 8.5 reflects a local attack vector (AV:L) with low privileges (PR:L), no user interaction (UI:N), low complexity (AC:L), and high confidentiality, integrity, and availability impact confined to the vulnerable system (VC:H/VI:H/VA:H, SC:N/SI:N/SA:N) - it is not remotely exploitable by an anonymous external client. No public exploit code and no confirmed active exploitation were identified at time of analysis; JPCERT/CC reported the issue and EG Secure Solutions and JVN have published advisories.
Command injection in the NetBird desktop client (versions 0.60.0 through 0.74.6) lets a malicious or compromised peer inject arbitrary OpenSSH directives - including command-executing ones - into the system-wide drop-in config of other clients' Linux, macOS, Windows and FreeBSD hosts. Exploitation requires the attacker to control an enrolled peer whose crafted FQDN or hostname containing a double quote is distributed to the victim, and the injected command runs only when a local user subsequently invokes ssh, so peer membership plus a local user action are both prerequisites rather than an unauthenticated internet-facing attack. NetBird 0.74.7 fixes the issue by rejecting peer names that are not valid DNS names; no public exploit identified at time of analysis, and the Android and iOS apps are not affected.
Unauthenticated OS command injection in the Collabora Online Built-in CODE Server Nextcloud app (richdocumentscode) before 26.4.303 allows remote attackers to execute arbitrary commands on the Nextcloud host via proxy.php, but only while the bundled CODE server process is not running - the window right after a restart, during startup, or following a crash. The CVSS 4.0 score is 9.2 and the CVE record attributes exploitation during September and October 2026, making this a genuine high-priority issue rather than a paper-tiger high-severity finding; the practical barrier is timing, not authentication, since the CVSS vectors carry PR:N/UI:N and an attacker can simply poll the endpoint until CODE is down. Standalone Collabora Online servers do not ship proxy.php and are explicitly not affected - the exposure is confined to Nextcloud instances with the richdocumentscode app, and the flaw is distinct from CVE-2025-66208.
OS command injection in the Samsung QA-Board backend allows unauthenticated remote attackers to execute arbitrary operating-system commands on the host running the service, yielding a CVSS 3.1 base score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) with full confidentiality, integrity and availability impact. The flaw, classified as CWE-78, affects all versions matched by the wildcard CPE cpe:2.3:a:samsung_open_source:qa-board:*:*:*:*:*:*:*:*, so no release line is known to be exempt. No public exploit code has been identified at time of analysis and the issue is not listed on CISA KEV, but the trivial remote/unauthenticated attack profile makes rapid patching the priority.
Command injection in the VNote note-taking application before version 4.3.0 lets an attacker-controlled note filename execute arbitrary commands as the victim user on Linux and macOS, but only when the victim opens that note and has configured an external program whose arguments are command-capable, such as vim/gvim's -c option. The root cause is SessionConfig::ExternalProgram::fetchCommand, which wraps the note path in double quotes and hands the whole string to QProcess::start(const QString&), so a double quote inside a POSIX filename terminates the quoted path and appends attacker-chosen arguments to the configured program. The vendor fixed the issue in 4.3.0; no public exploit identified at time of analysis, and the CVSS 3.1 vector (AV:L/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H, base 7.0) reflects full impact on confidentiality, integrity and availability once the narrow exploitation prerequisites are met.
Command injection in MultiTail 7.1.5 and earlier lets an attacker who can write content into a monitored log or stream execute arbitrary shell commands as the user running MultiTail, because matched text is concatenated into a command passed to sh -c with only partial escaping. Exploitation is gated by a deliberate, non-default configuration: the operator must have enabled the -ex/-eX execute-on-match feature with a filter regex, and the injected line must match that regex; because the CVSS vector is AV:L/PR:L/UI:R, this is a local, low-privileged, user-interaction-dependent issue with impact bounded by the privileges of the MultiTail process. Exploitation code is publicly available (CISA SSVC reports Exploitation: poc), but there is no evidence of confirmed active exploitation, and the assessed risk is moderate and internally consistent rather than alarming.
Command injection in MobSF v4.5.2 allows an attacker with access to the instance-and, for the /api/v1/scan path, a valid MobSF REST API key-to execute arbitrary shell commands via the /download_data/ endpoint, the /api/v1/scan REST API, or the Dynamic Analyzer ADB integration module (tests_common.py). The independent assessment scores this CVSS 8.8 (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H) rather than the 9.8 in the input because API-key authentication is enforced and the ADB path requires an active Android emulator or device. No public exploit identified at time of analysis, no CISA KEV listing, and EPSS is low at 0.71% (52nd percentile); MobSF is typically deployed internally, which reduces remote unauthenticated exposure.
OS command injection in Adobe Experience Manager 6.5 (through 6.5.25) and AEM 6.5 LTS (through SP3) allows an authenticated, low-privileged attacker to run arbitrary operating-system commands in the context of the current user, with changed scope (S:C) meaning the injected command executes outside the application's own security authority on the host. Rated CVSS 9.9 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H), the exploitation bar is low attack complexity with no user interaction; the only meaningful constraint is the need for a valid low-privilege AEM session, which is a very low bar in multi-user AEM author environments where content authors and other non-admin accounts are routinely issued. No public exploit code has been identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV); no EPSS score was supplied with the source data. This is a genuine high-priority vulnerability rather than a CVSS-inflated paper risk, and the specific vulnerable feature or injectable parameter is not named in the advisory summary, so patch validation and hunting should be prioritized on affected instances.
Command injection in Microsoft OneNote allows an unauthenticated remote attacker to execute code when a victim opens a specially crafted OneNote document. The issue is rated CVSS 8.8 (AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H) and affects Microsoft 365 Apps for Enterprise 16.0.1 per EUVD, with user interaction being the primary exploitation barrier. No public exploit was identified at time of analysis, and Microsoft has published an advisory and update path via the MSRC update guide.
Command injection in the Windows Shell lets remote, unauthenticated attackers bypass a Windows security feature and then execute arbitrary commands, yielding high impact to confidentiality, integrity, and availability. The attack is network-reachable with low complexity but requires victim interaction (CVSS UI:R) - the target must open or interact with attacker-supplied content such as a crafted file, shortcut, or link - which is why SSVC rates it non-automatable while still flagging total technical impact. The flaw affects Windows 10, Windows 11, and Windows Server 2012 R2 through 2025, plus Outlook.com, and is confirmed actively exploited (CISA KEV); given the KEV listing and the assessed high-end impact, this is a genuine top-tier patching priority rather than a high-CVSS-but-low-risk case.
Local privilege escalation in Windows Update enables an attacker who already holds a low-privileged local session to elevate to higher privileges by injecting special command elements into data processed by the Windows Update servicing component (CWE-77 command injection). This affects multiple Windows 11 releases (23H2, 24H2, 25H2, 26H1, 26H2) and Windows Server 2025, including Server Core, up to the fixed builds enumerated in Microsoft’s advisory, with a CVSS 3.1 score of 7.8 (AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H). Exploitation requires an existing local foothold and valid low-privileged credentials, and no public exploit has been identified at time of analysis; vendor patches are available.
Command injection in Visual Studio Code 1.0.0 through 1.141.0 allows an unauthenticated remote attacker to execute arbitrary code in the context of the editor when a developer opens attacker-controlled content such as a malicious workspace, folder, repository, or project configuration file. Exploitation depends on user interaction (a victim must open the crafted project), but is otherwise low-complexity with high confidentiality, integrity, and availability impact; no public exploit code has been identified at the time of analysis. Microsoft has released a vendor fix in Visual Studio Code 1.141.1.
OS command injection in KDE Dolphin 24.08.0 through 26.08.2 allows an attacker who can place a specially named directory on the victim's filesystem to run arbitrary commands with the victim's privileges when that folder's empty view background is double-clicked. The flaw lies in Dolphin substituting the folder path into the {path} placeholder of a shell command without neutralizing embedded double quotes and shell metacharacters, so a directory name such as one containing a quote plus `;`, `$()`, or backticks escapes the quoting in slotDoubleClickViewBackground. This is a genuine but interaction-gated local issue rather than a mass-exploitable remote flaw: exploitation requires the crafted directory to already exist (delivered via archive, network/share, USB, or cloned repository), an affected build running, and mandatory user interaction (the background double-click); no public exploit code was identified at time of analysis, and impact stays within the victim's user scope with no scope change.
OS command injection in rovr (a Python-based terminal file manager) through 0.10.3 on Windows allows a file name containing Windows shell metacharacters such as '&' to be passed unquoted to cmd.exe, executing arbitrary commands with the privileges of the user who opens that file via the open_file routine. Only Windows installations of rovr at or below 0.10.3 are affected; the vector is local and successful exploitation requires the victim to actively select and open an attacker-planted file, with the injected command running at the invoking user's existing privilege level (no elevation, no scope change). There is no CISA KEV listing and no public exploit identified at time of analysis; the CVSS 4.0 base score of 8.4 (High) reflects high confidentiality, integrity and availability impact, tempered by the local attack vector and mandatory user interaction.
OS command injection in Azure CycleCloud allows an authenticated local user with low privileges to execute arbitrary code on the affected host, but only if a victim performs a required user interaction. The CVSS 3.1 vector is AV:L/AC:L/PR:L/UI:R/S:U/C:H/I:H/A:H with a base score of 7.3, and the independent assessment rates this as a moderate-priority issue rather than a top-tier emergency. The available data provides no public exploit identified at time of analysis and no CISA KEV confirmation; Microsoft's MSRC update-guide entry is the provided remediation reference.
Authenticated OS command injection in the odt2spip plugin for SPIP before 4.4.0 allows author-level users to execute arbitrary commands as the web-server user by uploading a document with a crafted non-.odt filename through the document import form. The flaw survives sanitization in odt2spip_deplacer_fichier_upload() and LibreOffice::createCommand(), then triggers during LibreOffice conversion; exploitation requires a working LibreOffice backend and an author account. No public exploit was identified at time of analysis.
Authenticated users with workflow create/edit rights on self-hosted Heym deployments prior to 0.0.40 can escape the application's blacklist-based Python sandbox and execute arbitrary commands and file operations as the backend service account. The flaw lives in backend/app/services/python_tool_executor.py, where the allowlisted standard-library modules still expose OS, process, and filesystem primitives through public attributes, so tool source that passes _safe_import() and _validate_code_safety() is ultimately run through exec() with full application privileges (CVSS 8.8, CWE-78). Exploitation requires a low-privilege authenticated account (PR:L) with permission to create or edit a workflow containing an Agent custom Python tool; enabling self-registration makes that account trivially obtainable, though merely triggering a workflow from the portal does not grant code control. There is no public exploit identified at time of analysis and no confirmed active exploitation (not in CISA KEV), but the fix in 0.0.40 is available and the assessed risk is high for any multi-user deployment.
Unauthenticated remote command execution in Commvault CVFS, the file system component shipped with Commvault Grid (HyperScale X) deployments and exposed through Commvault Cloud, allows a network-reachable attacker to execute commands on the underlying system with high impact to confidentiality, integrity, and availability - Commvault scores it 9.3 under CVSS 4.0. Per the vendor CVSS vector (AV:N/AC:L/AT:N/PR:N/UI:N) the issue is remotely triggerable without credentials or user interaction, which makes any internet- or broadly reachable CVFS endpoint a priority exposure; the advisory does not name the specific feature, protocol endpoint, or configuration path that must be reached. No public exploit code and no confirmed active exploitation have been identified at time of analysis, and a fixed maintenance release is available from Commvault.
OS command injection in Wavlink WL-WN592AX6 firmware 250422 allows unauthenticated network attackers to execute arbitrary shell commands through the wmapd daemon's system function when they can reach the device management surface. The flaw is tracked as CVE-2026-109110 (CWE-78) with a 9.3 CVSS v4.0 score, and publicly available exploit code exists, though no confirmed active exploitation is indicated in the supplied data. Exploitation requires network access to the affected management interface; because this consumer range extender is typically LAN-facing, Internet exposure depends on WAN exposure or an attacker on the local/adjacent network.
Command injection in the Wavlink WL-WN530BE1-A wireless router (firmware 251117) allows an authenticated, low-privileged user of the web management interface to execute arbitrary shell commands on the device via the 'Type' parameter of /cgi-bin/uploadVpn.cgi, which is passed unsanitized into a popen() call. Because the flaw requires only a valid session (CVSS PR:L) with no user interaction (UI:N) and generally only LAN-side reachability to the management interface, the practical barrier is low: SOHO devices frequently ship with default or weak admin credentials. Publicly available exploit code exists (a published GitHub gist), but no confirmed active exploitation (CISA KEV) has been reported, and no vendor-released patch version is identified in the available data.
Command injection in the backup function of Teltonika Networks RUTOS and TSWOS allows an attacker who already holds high-privilege administrative access on an adjacent network segment to execute arbitrary OS commands on the device. Affected builds are RUTOS 7.22 before 7.24.6 and 7.25 before 7.25.2 on RUT, RUTX, RUTM, RUTC, ATRM, CAP, OXA, DAP, TRB, OTD and EBG devices, RUTOS 7.22 before 7.25.2 on TAP devices, and TSWOS 1.9 before 1.10.4 on TSW and SWM devices. No public exploit has been identified at time of analysis, and the assessed prerequisites (PR:H for administrator credentials plus AV:A for management-plane adjacency) mean exploitation is not reachable by an arbitrary internet attacker; patched firmware is available.
Remote OS command execution in IBM DataStage on Cloud Pak for Data 5.4.0.0 allows an authenticated attacker holding only low-privileged credentials to run arbitrary operating system commands on the underlying DataStage host, with high impact to confidentiality, integrity, and availability (CVSS 8.8, AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H). The flaw is an OS command injection (CWE-78) in which specially crafted input is passed to a system command without proper neutralization, so no user interaction is required and a single crafted request over the network is sufficient once valid access is obtained. Exploitation is gated by the need for existing credentials on the instance, and there is no public exploit code identified at time of analysis; IBM has published a fix via support node 7288649.
Authenticated remote code execution in IBM DataStage on Cloud Pak for Data 5.4.0.0 allows an attacker who already holds valid low-privilege credentials (CVSS PR:L, AV:N, UI:N) to run arbitrary operating system commands on the underlying host, resulting in high confidentiality, integrity, and availability impact (CVSS 8.8). The root cause is improper neutralization of special elements used in an OS command (CWE-94), meaning user-supplied input reaches a shell command without adequate sanitization. No public exploit code has been identified at time of analysis and the flaw is not confirmed actively exploited (not in CISA KEV), but a vendor patch is available per IBM's advisory, so unpatched 5.4.0.0 deployments exposed to multi-tenant users or broadly provisioned accounts should be prioritized for upgrade.
Command injection in the IBM Maximo Application Suite 'Location Service for Esri' component (versions 9.0, 9.1 and 9.2) allows a highly privileged attacker to execute arbitrary shell commands through a Kubernetes operator task, compromising confidentiality, integrity and availability beyond the component's own security boundary (CVSS 8.7). Exploitation is access-gated rather than trivial: the optional Esri location component must be deployed and enabled, and the attacker must already hold high privileges in the MAS environment (PR:H), but once those two conditions hold there is no user-interaction requirement and attack complexity is low, so injection is straightforward. No public exploit code and no confirmed active exploitation were identified at time of analysis.
Unauthenticated command injection in Edimax BR-6208AC firmware version 1.32 allows a remote attacker to execute arbitrary operating-system commands on the router by supplying crafted input through the readE2P argument of the /goform/formWlanMP endpoint. The issue is rated critical (CVSS 4.0 base 9.3; an independent assessment reaches the equivalent CVSS 3.1 vector AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) and publicly available exploit code exists, although there is no confirmation of active exploitation (device is not in CISA KEV). The vendor was contacted early and did not respond, and the product is no longer supported, so the decisive practical concern is permanent exposure: any BR-6208AC 1.32 whose web management interface is reachable over the WAN or from an untrusted network segment remains exploitable indefinitely.
Unauthenticated command injection in Edimax BR-6208AC firmware version 1.32 lets remote attackers execute arbitrary shell commands on the router by submitting crafted input to the tftpsIP parameter of the /goform/formAutoFWupgrade handler, which normally accepts a TFTP server address for firmware fetch operations. The vendor was contacted about the disclosure but did not respond, and the device is an end-of-life product no longer supported by the maintainer, so no vendor-released patch is expected; publicly available exploit code exists (a GitHub proof-of-concept under an IoT vulnerability repository) and the CVSS 4.0 score of 9.3 reflects a full confidentiality, integrity, and availability impact. The vulnerability is not listed in CISA KEV (no confirmed active exploitation reported), but pre-auth reachability of the HTTP management interface - particularly if WAN administration is enabled - makes this a high-priority exposure for anyone still running the device.
Authenticated content editors on Grav CMS sites running the Git Sync plugin versions 1.0.2 through 3.4.12 can execute arbitrary OS commands as the web-server account when the plugin's gravuser or gravfull author mode is enabled, by planting shell substitution payloads in their username, full name, or email profile fields that are later interpolated into a git commit --author call and reach PHP exec(). The vulnerability is rated 8.7 (CVSS 4.0) and maps to CWE-78; the assessed vector (CVSS 3.1 AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H) confirms that exploitation requires authenticated, low-privileged editor access but no victim interaction, with the additional gating condition that author mode must be explicitly turned on and must not be left at its default posture. No public exploit code has been identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV), so it is a credible, prioritizable bug rather than a drop-everything emergency - though any site with untrusted or broadly provisioned editor accounts and author mode enabled should treat it as an immediate command-execution risk.
Authenticated OS command injection in the Linksys E1700 router's embedded web management interface (firmware 1.0.0.4.003) lets a remote attacker with low-privilege credentials execute arbitrary operating-system commands on the device by manipulating the _6rd_prefix and _6rd_prefix_len parameters handled by setLinksysIpv6 at /goform/setLinksysIpv6. The flaw scores 8.6 under CVSS 4.0 (AV:N/AC:L/AT:N/PR:L/UI:N, E:P) and 8.8 under CVSS 3.1 (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), reflecting full confidentiality, integrity and availability impact on the router. Publicly available exploit code exists - including a published proof-of-concept write-up referenced from the CVE record - but there is no confirmed actively exploited (CISA KEV) status, and no vendor patch has been released because Linksys did not respond to the disclosure; practical exploitation requires reachability of the HTTP management interface plus valid (or default/weak) credentials, which on default deployments is typically LAN-side.
OS command injection in the Linksys E1700 router running firmware 1.0.0.4.003 lets an attacker execute arbitrary shell commands on the device by sending a crafted request to the /goform/setDMZ endpoint with a manipulated MAC_addr parameter. Exploitation requires network reachability to the web management interface (LAN-side by default, or internet-facing if remote administration is enabled) plus at least low-privilege authenticated access, and needs no user interaction; the vendor did not respond to the disclosure, so no patch is available. Publicly available exploit code exists (a proof-of-concept is published on GitHub), which raises the likelihood of opportunistic attacks against exposed devices.
OS command injection in Linksys E1700 firmware 1.0.0.4.003 allows authenticated low-privilege attackers to execute arbitrary OS commands on the device via crafted input to the static route parameters in /goform/addStaticRoute. Publicly available exploit code exists, increasing the risk of exploitation; no vendor patch has been released as Linksys did not respond to the disclosure. The vulnerability requires network access to the management interface and valid credentials, but successful exploitation grants full control of the router.
OS command injection in the Ministry of Justice General Directorate of Information Technology's UYAP Document Editor before version 5.4.20 allows an attacker to execute arbitrary operating-system commands on a victim's machine when a crafted document is opened in the application. The issue is local in nature (AV:L) and requires user interaction (UI:R) - the victim must open a malicious file - but no authentication or privileges are needed (PR:N), and successful exploitation yields high confidentiality, integrity, and availability impact, giving a base score of 7.8 (HIGH) on the CVSS 3.1 vector AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H. The flaw was reported by TR-CERT; no public exploit code has been identified and the issue is not listed in CISA KEV, so the practical risk is driven by phishing-style delivery of weaponized-looking documents rather than by remote scanning or automated mass exploitation.
OS command injection in mac-shell-mcp before 2.0.0 lets a connected MCP client - including a legitimate client driven by a malicious or prompt-injected LLM - escape the tool's command whitelist by appending shell metacharacters to the arguments of an otherwise permitted command such as pwd, causing forbidden commands to run as the server user without approval. Exploitation requires an existing MCP client session to the local server (CVSS 4.0 vector AV:L/PR:L/UI:N, mirroring the independent CVSS 3.1 assessment AV:L/PR:L/UI:N with high confidentiality, integrity and availability impact), so this is an authenticated, locally-scoped flaw rather than an internet-facing unauthenticated one, and impact is bounded by the privileges of the account the server runs under. No public exploit code has been identified at time of analysis and the issue is not confirmed actively exploited (CISA KEV), but a vendor fix and upstream patch commit are available; EPSS data was not provided for this CVE.
OS command injection in Tuleap Enterprise Edition versions 17.0 through 17.5 allows an authenticated, low-privilege user to execute arbitrary commands on the underlying server, delivering a full compromise of confidentiality, integrity, and availability (CVSS 3.1 base 8.8, AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H). Because the assessed vector specifies PR:L, exploitation requires a valid account on the target instance - a registered project or site user is sufficient - and no user interaction is needed once that access is obtained, so unauthenticated attackers cannot trigger it directly. No public exploit code was identified at time of analysis and the CVE is not listed in CISA KEV; no EPSS score was supplied in the source data. Practical risk is therefore concentrated on internet-reachable Tuleap Enterprise Edition instances in the 17.0-17.5 range that allow self-registration or broad account provisioning, since such deployments effectively lower the credential barrier to exploitation.
Unauthenticated OS command injection in the Web Management Backend of TRENDnet TEW-751DR, TEW-752DRU, and TEW733GR routers (firmware 1.03B01/1.03B03) allows remote attackers to execute arbitrary shell commands on the device by sending crafted input in the 'service' argument handled by the fwrite function of pigwidgeon.cgi. Per the CVSS vector (PR:N/UI:N/AC:L) no authentication, user interaction, or non-default configuration is required; the principal limiting factor is reachability of the management interface, which is LAN-side by default and only exposed over WAN where remote administration is enabled or the router is internet-facing. Publicly available exploit code exists (a GitHub Proof-of-Concept targeting these models), so exposed units should be treated as compromisable with minimal attacker effort; EPSS and CISA KEV data were not provided, so confirmed active exploitation cannot be asserted. Vendor has confirmed the affected models reached end-of-life in 2011 and will not be remediated by a firmware fix.
OS command injection in GNU Emacs's TRAMP remote-file library (Emacs releases before 31.2, TRAMP through 2.8.2) lets an attacker who can get a crafted filename in front of a user execute arbitrary shell commands with that user's privileges, because tramp-user-regexp relies on an incomplete denylist of dangerous shell characters. This is a follow-on defect to CVE-2026-79992, whose fix did not cover all dangerous inputs. Exploitation is local, requires user interaction (the victim must open, complete, or otherwise have Emacs/TRAMP evaluate the malicious path), and only affects installations that use TRAMP remote-file functionality; no public exploit code and no confirmed active exploitation were identified at time of analysis.
Databasement instances before 1.8.2 allow an authenticated user to achieve remote code execution by injecting path-redirecting client options (such as mariadb-dump --result-file, -r, --tab, pg_dump --file, mongodump --out, or redis-cli --rdb/--eval) into the arguments passed to the backup dump binary. Shell quoting prevents classic OS command injection, but argument injection alone is sufficient: by directing the dump output to a web-served, PHP-executing path such as /app/public/index.php, the attacker writes code that the web server later executes. Authentication is required (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, base 8.8), publicly available exploit code exists, and success depends on the chosen output path being both web-accessible and interpretable as code; no evidence of confirmed active exploitation was provided.
OS command injection in pbi-cli 3.10.1 through 3.12.0 allows a local attacker to execute arbitrary commands with victim privileges when the victim opens a crafted Power BI .pbip project from a space-free filesystem path containing an '&' character. The flaw resides in desktop_sync.py, which passes unquoted paths to 'cmd /c start' during report write or reload, and exploitation requires user interaction (UI:P) with the crafted path. No public exploit code or active exploitation (CISA KEV) has been identified at time of analysis; CVSS 4.0 score is 7.3 (High), and an independent CVSS 3.1 assessment yields 7.8 (High).