The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes source code from local repositories, creating a surface where malicious instructions hidden in code comments or strings could be analyzed and potentially executed by the agent.
Ingestion points: The scripts/scan_sqli.py script reads the contents of various source file types within the target directory.
Boundary markers: The skill does not currently employ specific delimiters or clear instructions to the agent to treat findings as untrusted data.
Capability inventory: The skill is configured with Read, Bash, Glob, and Grep capabilities, and the scanner performs filesystem read operations.
Sanitization: Code snippets matched by the regex scanner are presented to the agent without specific filtering to neutralize potential embedded instructions.

detecting-sql-injection-patterns