webapp-sqlmap

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt repeatedly shows embedding sensitive values verbatim into commands and headers (e.g., --cookie="sessionid=abc123", Authorization: Bearer token123, --auth-cred="admin:password"), which requires the agent to handle and output secrets directly and thus creates an exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The document is a high-risk, dual-use offensive tool guide that explicitly documents techniques to exfiltrate data, steal credentials, obtain remote shells, write webshells/files to servers, and evade defenses—features that can be (and often are) abused for unauthorized compromise despite the authorization caveat.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The included CI template runs a runtime command that fetches and executes remote code via "curl -s https://raw.githubusercontent.com/aquasecurity/tfsec/master/scripts/install_linux.sh | bash", which is a high-confidence external dependency that executes remote code during skill runtime.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The prompt instructs use of sqlmap to perform invasive actions (file-read, file-write, --os-cmd, --os-shell, writing web shells) that modify system files and execute commands on target hosts and could directly compromise the agent's host if targeted (e.g., localhost), so it encourages state-compromising behavior.