🧪

Audit-Verlauf

webapp-testing - 5 Audits

Audit-Version 5

Neueste Niedriges Risiko

Jan 16, 2026, 05:08 PM

Legitimate web testing toolkit using Playwright. Static findings are false positives: LICENSE.txt cryptographic warnings are standard legal text, subprocess usage is user-controlled for local dev servers, markdown backticks in SKILL.md are documentation formatting. No data exfiltration, credential theft, or persistence mechanisms detected.

7
Gescannte Dateien
814
Analysierte Zeilen
5
befunde
claude
Auditiert von
Probleme mit mittlerem Risiko (1)
Shell execution with user-controlled commands
with_server.py uses subprocess.Popen with shell=True to execute server commands provided via CLI. This is intentional for supporting compound shell commands. Commands are user-provided for local development.

Risikofaktoren

⚡ Enthält Skripte (1)
🌐 Netzwerkzugriff (1)
📁 Dateisystemzugriff (1)
⚙️ Externe Befehle (1)

Audit-Version 4

Niedriges Risiko

Jan 16, 2026, 05:08 PM

Legitimate web testing toolkit using Playwright. Static findings are false positives: LICENSE.txt cryptographic warnings are standard legal text, subprocess usage is user-controlled for local dev servers, markdown backticks in SKILL.md are documentation formatting. No data exfiltration, credential theft, or persistence mechanisms detected.

7
Gescannte Dateien
814
Analysierte Zeilen
5
befunde
claude
Auditiert von
Probleme mit mittlerem Risiko (1)
Shell execution with user-controlled commands
with_server.py uses subprocess.Popen with shell=True to execute server commands provided via CLI. This is intentional for supporting compound shell commands. Commands are user-provided for local development.

Risikofaktoren

⚡ Enthält Skripte (1)
🌐 Netzwerkzugriff (1)
📁 Dateisystemzugriff (1)
⚙️ Externe Befehle (1)

Audit-Version 3

Mittleres Risiko

Jan 10, 2026, 10:40 AM

This skill provides legitimate web testing functionality using Playwright. It includes a server management script that executes user-provided shell commands for starting local development servers. The shell=True usage in subprocess is a known security concern but is intentional for supporting compound shell commands like 'cd backend && python server.py'. No credential theft, data exfiltration, or persistence mechanisms detected.

6
Gescannte Dateien
456
Analysierte Zeilen
5
befunde
claude
Auditiert von
Probleme mit mittlerem Risiko (1)
Shell command execution with user input
The with_server.py script uses subprocess.Popen with shell=True to execute server commands provided via command-line arguments. While this is intentional for supporting compound shell commands (e.g., 'cd backend && python server.py'), shell=True combined with user input is a known security concern. An attacker who controls the command arguments could potentially execute arbitrary commands. Relevant code (scripts/with_server.py:69-74): ```python process = subprocess.Popen( server['cmd'], shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE ) ``` This is mitigated by: (1) commands come from the user's own arguments, (2) the skill is designed for local development server management, (3) no remote execution vector exists.

Audit-Version 2

Mittleres Risiko

Jan 10, 2026, 10:40 AM

This skill provides legitimate web testing functionality using Playwright. It includes a server management script that executes user-provided shell commands for starting local development servers. The shell=True usage in subprocess is a known security concern but is intentional for supporting compound shell commands like 'cd backend && python server.py'. No credential theft, data exfiltration, or persistence mechanisms detected.

6
Gescannte Dateien
456
Analysierte Zeilen
5
befunde
claude
Auditiert von
Probleme mit mittlerem Risiko (1)
Shell command execution with user input
The with_server.py script uses subprocess.Popen with shell=True to execute server commands provided via command-line arguments. While this is intentional for supporting compound shell commands (e.g., 'cd backend && python server.py'), shell=True combined with user input is a known security concern. An attacker who controls the command arguments could potentially execute arbitrary commands. Relevant code (scripts/with_server.py:69-74): ```python process = subprocess.Popen( server['cmd'], shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE ) ``` This is mitigated by: (1) commands come from the user's own arguments, (2) the skill is designed for local development server management, (3) no remote execution vector exists.

Audit-Version 1

Mittleres Risiko

Jan 10, 2026, 10:40 AM

This skill provides legitimate web testing functionality using Playwright. It includes a server management script that executes user-provided shell commands for starting local development servers. The shell=True usage in subprocess is a known security concern but is intentional for supporting compound shell commands like 'cd backend && python server.py'. No credential theft, data exfiltration, or persistence mechanisms detected.

6
Gescannte Dateien
456
Analysierte Zeilen
5
befunde
claude
Auditiert von
Probleme mit mittlerem Risiko (1)
Shell command execution with user input
The with_server.py script uses subprocess.Popen with shell=True to execute server commands provided via command-line arguments. While this is intentional for supporting compound shell commands (e.g., 'cd backend && python server.py'), shell=True combined with user input is a known security concern. An attacker who controls the command arguments could potentially execute arbitrary commands. Relevant code (scripts/with_server.py:69-74): ```python process = subprocess.Popen( server['cmd'], shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE ) ``` This is mitigated by: (1) commands come from the user's own arguments, (2) the skill is designed for local development server management, (3) no remote execution vector exists.