The Agent Skills Directory

[SAFE]: The skill is a collection of documentation and code templates providing infrastructure for other skills. It does not perform any automated malicious actions or possess hidden capabilities. It emphasizes security best practices like safe command usage and least-privilege module loading.
[COMMAND_EXECUTION]: The modules include various shell command examples for git operations, system information gathering, and build tool interactions. These serve as instructional templates and do not represent autonomous or unauthorized command execution.
[EXTERNAL_DOWNLOADS]: The skill references standard development tools and libraries (e.g., cargo-audit, pytest, py-spy, clinic.js) to be used via official package managers. These are documented for legitimate development and analysis purposes.
[DATA_EXFILTRATION]: No network-based exfiltration was found. The skill's telemetry and caching patterns operate entirely on the local file system for session state management.
[INDIRECT_PROMPT_INJECTION]: The skill addresses potential injection risks by providing patterns for robust data processing, including size-based chunking and structured document analysis, which help maintain clear context boundaries.

progressive-loading