The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several command-line tools including git, gh (GitHub CLI), go test, and make. These are used to manage repository state, run verification suites, and push code changes. These operations are standard for a development-focused skill.
[EXTERNAL_DOWNLOADS]: The skill retrieves pull request metadata and review comments from GitHub's official API and CLI. These network operations are directed at a well-known, trusted service to facilitate the primary workflow.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it ingests untrusted data from pull request reviews and comments which could contain malicious instructions.
Ingestion points: Pull request review bodies, inline comments, and issue comments are fetched via gh pr view and the GitHub API (SKILL.md).
Boundary markers: Absent. There are no explicit instructions or delimiters used to separate the untrusted PR content from the skill's own operational instructions.
Capability inventory: The skill has the ability to modify code, execute arbitrary shell commands via make and go test, and push changes to a remote repository via git push (SKILL.md).
Sanitization: None. The skill relies on the agent's ability to triage findings based on priority rather than technical sanitization or instruction filtering.

pr-finish