The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the gh and git command-line tools to perform legitimate issue and repository management tasks. These include administrative actions like locking threads, pinning issues, and creating linked development branches.
[DYNAMIC_CONTEXT_INJECTION]: Uses the ! command syntax in SKILL.md to run git remote get-url origin and gh issue list when the skill is loaded. These commands are used solely to provide necessary context about the current repository and its open issues to the agent.
[PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface by ingesting data from GitHub issues (titles and metadata).
Ingestion points: The Context section and validation steps in SKILL.md read external data from GitHub via the gh CLI.
Boundary markers: No specific boundary markers are employed to isolate the ingested issue content from the agent's instructions.
Capability inventory: The skill can modify issues, create branches, and execute arbitrary GitHub API calls via the gh tool.
Sanitization: External content is not sanitized before being presented to the agent, which is standard for this type of development tool.

git-issue-manage