The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from GitHub pull requests, including code diffs, PR descriptions, and existing review comments.
Ingestion points: Pull request metadata and review threads are fetched via gh pr view and gh api graphql (Steps 2 and 3). The pull request code is fetched via git fetch and checked out into a worktree (Step 5).
Boundary markers: The instructions do not specify the use of delimiters or explicit 'ignore embedded instructions' warnings when passing external content to the /code-review and writing-style skills.
Capability inventory: The skill can post reviews to GitHub, switch the active authenticated user, and create/delete local worktrees.
Sanitization: There is no explicit mention of sanitizing or escaping the content retrieved from GitHub before it is processed by the AI to determine the review event or draft comments.
[COMMAND_EXECUTION]: The skill executes several shell commands to manage the environment and interact with GitHub.
Tooling: Uses git for worktree management and gh (GitHub CLI) for authentication, API interactions, and PR viewing.
Authentication Management: In Step 10, the skill uses gh auth switch --user to change the local environment's authenticated user. While it attempts to switch back, an interrupted execution could leave the local environment set to the secondary account.

review-comments