The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted data from GitHub issue titles and bodies to perform heuristic scoring and automated labeling.
Ingestion points: scripts/label_agent_issues.py fetches issue titles and bodies using the gh issue list command.
Boundary markers: No boundary markers or instructions to ignore embedded commands are present when processing the issue data.
Capability inventory: The skill possesses the capability to create and apply labels, post comments with mentions, and assign milestones via the gh CLI.
Sanitization: There is no evidence of sanitization or escaping of issue content before it is used in heuristic calculations or interpolated into comments.
[COMMAND_EXECUTION]: The script performs automated actions using the gh and git command-line interfaces.
Evidence: scripts/label_agent_issues.py uses subprocess.run to execute commands such as gh label create, gh issue edit, and gh issue comment.
Context: While the script uses the safer list-of-arguments format for subprocess.run, the arguments are partially derived from untrusted issue data, which could lead to unexpected CLI behavior if not properly handled by the underlying tool.

github-label-agent-issues