The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from GitHub PRs to make logic decisions.\n
Ingestion points: The skill retrieves the PR title, body, and code changes using gh pr view and gh pr diff in SKILL.md.\n
Boundary markers: There are no explicit delimiters or instructions to ignore commands potentially embedded within the PR content.\n
Capability inventory: The skill has significant capabilities, including merging code to the main branch (gh pr merge), posting comments/reviews (gh pr review, gh pr comment), and creating new releases (gh release create) in SKILL.md.\n
Sanitization: There is no logic present to sanitize the PR body or diff text before the agent evaluates it for merge criteria.\n- [COMMAND_EXECUTION]: The skill relies on the GitHub CLI (gh) to perform repository management. While these are legitimate tasks for this skill's purpose, the tool's execution flow is triggered by external PR data.

review-pr