Security audit
Security checks for vulnerabilities and agentic risk
This is a straightforward data-analysis skill whose file-writing examples are disclosed and purpose-aligned, with no evidence of hidden execution, persistence, credential access, or exfiltration.
Before using this skill, run analysis only on datasets you intend the agent to process and direct outputs to an isolated folder to avoid overwriting files or creating sensitive derived reports in shared locations.
The activation guidance uses generic phrases such as 'User wants to analyze a dataset' and 'Generate charts and visualizations' without narrowing context, explicit trigger phrases, or exclusion conditions. These descriptions overlap with common analytics requests and make it unclear when this skill should or should not activate.
The skill repeatedly shows commands that write reports, charts, and transformed datasets to disk, but it does not warn that execution will create or overwrite files at the provided output paths. In agent contexts, missing disclosure about filesystem side effects can lead to unintended data exposure, clobbering existing files, or writing sensitive derived artifacts to insecure locations.
