The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the official gh (GitHub CLI) and git command-line tools as documented in references/trackers/github.md. These tools are used to retrieve repository information, check authentication, and manage issue life cycles (view, create, and close).
[EXTERNAL_DOWNLOADS]: The skill interacts with GitHub's official infrastructure via the gh tool to process issues. These network operations are necessary for the skill's stated purpose and target a well-known, trusted service.
[SAFE]: The skill includes a mandatory confirmation step ("Confirm once, then file") where the user must approve the ticket's title, body, and labels before the agent executes the creation command. This provides a human-in-the-loop safety check against accidental or unwanted operations.
[PROMPT_INJECTION]: The skill has a potential surface for indirect prompt injection as it ingests content from untrusted project files like CONTRIBUTING.md and AGENTS.md (documented in references/repository-conventions.md).
Ingestion points: Reads CONTRIBUTING.md, .github/CONTRIBUTING.md, and AGENTS.md to identify project-specific ticket conventions.
Boundary markers: Absent; the skill reads file content and applies instructions found within to the current task.
Capability inventory: The skill has the capability to execute gh issue create and gh issue close via the references/trackers/github.md reference file.
Sanitization: Absent; however, the impact is mitigated by the mandatory user confirmation step before any action is taken.

open-ticket