The Agent Skills Directory

[COMMAND_EXECUTION]: The skill interpolates the $ARGUMENTS variable directly into multiple shell command blocks, including static analysis and test execution steps. This construction is vulnerable to shell command injection; if the input contains characters like ;, &, or |, it could lead to the execution of unintended and potentially malicious commands on the system.
[EXTERNAL_DOWNLOADS]: The skill uses uv run to fetch and execute prek, ruff, and pytest. While ruff and pytest are standard tools, prek is not a widely recognized package. Executing code from unverifiable or niche packages introduces a risk of supply chain attacks.
[INDIRECT_PROMPT_INJECTION]: The skill processes project files and paths provided as arguments. It lack boundary markers or explicit instructions to ignore embedded prompts within analyzed code.
Ingestion points: File paths and source code accessed via $ARGUMENTS.
Capability inventory: Shell command execution.
Boundary markers: Absent.
Sanitization: Absent.

cleanup