The Agent Skills Directory

[SAFE]: The skill serves as a security auditing guide. It describes common SQL injection patterns, detection techniques (error-based, time-based, boolean-based), and database-specific payloads (MySQL, PostgreSQL, MSSQL, Oracle).
[COMMAND_EXECUTION]: While the skill contains shell commands (e.g., grep, curl, mitmdump), these are provided as examples for a user to perform security analysis on their own traffic (log.txt). There is no automated execution of dangerous commands or hidden remote code execution. The use of $ARGUMENTS is in a descriptive context rather than being piped directly into an execution sink.
[DATA_EXFILTRATION]: There is no evidence of unauthorized data exfiltration. The network commands (curl) are presented as methodology for testing a user's own target application for vulnerabilities.

mitm-find-sqli