The Agent Skills Directory

[SAFE]: The skill uses the official GitHub CLI (gh) for its operations. References to well-known services such as GitHub and its official CLI are considered safe practices.
[COMMAND_EXECUTION]: The skill executes standard GitHub CLI commands (gh pr view, gh pr diff, gh pr review) to perform its stated function. These operations are transparent, part of a guided multi-phase process, and directly support the primary purpose of the skill.
[PROMPT_INJECTION]: There is a surface for indirect prompt injection as the skill ingests untrusted data from pull request titles, descriptions, and code diffs.
Ingestion points: Data enters the context via gh pr view, gh pr diff, and gh issue view commands as described in Phase 1 of SKILL.md.
Boundary markers: Absent; the instructions do not explicitly provide delimiters or warnings to ignore instructions embedded within the PR content.
Capability inventory: The skill has the ability to write data back to GitHub using the gh pr review command in Phase 5.
Sanitization: Risk is mitigated by a mandatory human-in-the-loop checkpoint in Phase 4, where the user must review and approve the draft before submission.

review-pr