pentest-skills-framework

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests outsider-authored free text from the target’s HTTP responses (e.g., exploit-sqli/scripts/sqli_tester.py and exploit-xss/scripts/xss_tester.py use requests.get(...).text and then the agent may incorporate those response bodies into its LLM context for analysis/reporting), which is not authored by the operating user.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill documentation explicitly instructs running sudo commands (e.g., sudo apt install, sudo nmap -sS) and modifying user groups with sudo usermod -aG, which require elevated privileges and change the host system state.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)