The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/warp_compile_probe.py script uses subprocess.run() to execute a user-specified command. This is the core functionality of the skill, designed to profile the JIT compilation behavior of a specific Warp-based application. The script uses an isolated environment by overriding cache paths and does not require elevated privileges.
[REMOTE_CODE_EXECUTION]: While the skill involves executing code, it is restricted to the local command provided by the user for profiling purposes. There are no patterns of fetching and executing scripts from remote or untrusted servers.
[DATA_EXFILTRATION]: The skill instructions and scripts do not access sensitive file paths (like .ssh or .aws) or perform network operations to external domains. Cache directories are created in the system temporary directory and managed locally.
[PROMPT_INJECTION]: The skill instructions are purely technical and do not contain patterns intended to bypass AI safety filters or override system instructions.
[PERSISTENCE_MECHANISMS]: No attempts to modify shell profiles, cron jobs, or registry keys were detected. The skill uses atexit hooks in Python to clean up temporary directories after execution.

warp-compile-time-optimizer