The Agent Skills Directory

[SAFE]: The skill performs legitimate documentation and configuration tasks using local templates and scripts. It facilitates the definition of Service Level Objectives (SLOs) and Service Level Indicators (SLIs) within a project's documentation structure.
[COMMAND_EXECUTION]: The skill utilizes common shell commands such as mkdir, rsync, and cp to manage file structures and copy internal templates. It also invokes local Python scripts (scripts/build_docs.py, .github/actions/gate/run.py) for document validation and linting. These operations are restricted to the local filesystem and the plugin's own resources.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied inputs, including service IDs, user journey descriptions, and metric names. While this data is interpolated into documentation templates, the resulting files are stored locally, and the skill does not expose this data to external network endpoints or high-privilege system operations, making the injection surface minimal and low-risk.

slo