The Agent Skills Directory

[COMMAND_EXECUTION]: The Python scripts scripts/capture_sim_memgraph.py and scripts/summarize_memgraph.py utilize the subprocess.run function to invoke Apple developer tools such as xcrun, simctl, and leaks. These commands are executed using list-based arguments, which is a secure practice that avoids shell injection vulnerabilities.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data from .memgraph files and the output of the leaks utility. The parser in scripts/summarize_memgraph.py employs strict regular expression patterns to extract memory addresses and object sizes, ensuring that the data is handled safely before being presented to the agent.
[DATA_EXFILTRATION]: All operations, including the capture and analysis of memory graphs, are performed locally. The skill outputs artifacts to user-defined directories on the local file system and does not contain any code for network communication or external data transmission.

ios-memgraph-analysis