The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using the GitHub CLI (gh).
It fetches issue data using gh issue view and updates issue content using gh issue edit.
The commands are scoped to the elastic/kibana repository, which aligns with the skill author's identity ('elastic').
The skill uses a shell heredoc (EOF) in Step 6 to safely pass the new issue body to the CLI, which is a best practice to prevent command injection from the generated content.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from GitHub issues.
Ingestion points: External data enters the agent context via the gh issue view command in SKILL.md (Step 1), which retrieves the title and body of a specified issue.
Boundary markers: The instructions do not define clear boundary markers or provide 'ignore embedded instructions' warnings when the agent reformats the issue content in Step 3.
Capability inventory: The agent possesses write capabilities via the gh issue edit command in SKILL.md (Step 6), allowing it to modify the repository's content.
Sanitization: There is no explicit sanitization or filtering of the fetched issue content before it is used to generate the new body. However, the skill mandates a human review and confirmation step (Step 5) before any write operation is executed, which serves as a significant manual mitigation.

gh-enhance-issue