The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external application source code, which serves as an ingestion point for untrusted data. Maliciously crafted source code containing instructions could potentially influence the agent's behavior.
Ingestion points: scripts/analyze_memory.py performs recursive directory scanning and file analysis on target paths.
Boundary markers: The skill lacks explicit instructions to treat code content strictly as data or to ignore embedded instructions.
Capability inventory: The skill utilizes the Bash tool and possesses the ability to write and execute files via provided scripts.
Sanitization: No sanitization or validation of the analyzed code content is performed before processing.
[DYNAMIC_EXECUTION]: The script scripts/generate_report.py provides functionality to generate shell scripts. It takes a template string, writes it to a file with a .sh extension, and applies chmod 0o755 to make the file executable. This mechanism allows the creation and staging of arbitrary executable code.
[COMMAND_EXECUTION]: The skill configuration in SKILL.md explicitly allows the use of the Bash tool for profiling and memory management tasks, providing an interface for shell command execution within the scoped toolset.

detecting-memory-leaks