The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external datasets in CSV or JSON format via the scripts/decompose.py utility. The potential surface for injection through these files is mitigated by strict numeric validation within the script, which refuses to process non-finite or unparseable values, ensuring data is handled as statistical input rather than instructions.
[EXTERNAL_DOWNLOADS]: The documentation references various academic sources and research papers for methodological grounding. These are provided as static informational references and do not trigger automated downloads, remote code execution, or package installations.
[COMMAND_EXECUTION]: The provided Python script is limited to local data processing and statistical calculation. It relies exclusively on the Python standard library and does not invoke shell commands, spawn untrusted subprocesses, or perform network operations.
[DATA_EXPOSURE]: The analysis utility reads local files specified by the user to generate statistical summaries. It does not access sensitive system directories, environment variables, or hardcoded credentials.

data-verification