The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute python3 scripts that are dynamically generated based on user input.
[REMOTE_CODE_EXECUTION]: In Step 3 and Step 4, placeholders such as OUTPUT_DIR, YEAR, and DOWNLOADED_FILES are replaced with user-supplied values within Python code blocks. Because these values are interpolated directly into string literals without sanitization or escaping instructions, it is possible for a user to break out of the string and execute arbitrary Python code.
[REMOTE_CODE_EXECUTION]: The skill performs dynamic code generation by incorporating untrusted external input into executable scripts via string replacement, which constitutes a high-risk pattern when processed by the agent without safety constraints.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface with the following characteristics:
Ingestion points: Command line arguments ($@) and natural language descriptions parsed in Step 1.
Boundary markers: Absent for the interpolated variables in the Python script templates.
Capability inventory: File system read/write, Bash shell access, and Python code execution.
Sanitization: Absent; the skill does not instruct the agent to escape or validate user-supplied strings before script interpolation.

download-data