The Agent Skills Directory

[SAFE]: The skill is a collection of templates and instructions for writing PR bodies. It does not contain executable code, scripts, or remote dependencies.
[SAFE]: All external links in the documentation refer to legitimate public GitHub repositories for attribution purposes.
[SAFE]: Mentions of the GitHub CLI (gh) occur within code blocks as examples for standard development workflows, such as attaching images to PRs. These are informative and do not initiate unauthorized network operations or exfiltrate sensitive data.
[INDIRECT_PROMPT_INJECTION]: The skill functions by ingesting project context and user-supplied diffs to generate summaries. While this represents a data ingestion surface, the skill lacks any dangerous capabilities (such as arbitrary command execution or external data transmission) that would allow for exploitation via indirect injection.
[SAFE]: The YAML frontmatter and agent configuration are consistent with the skill's stated purpose of assisting with documentation tasks.

pr