The Agent Skills Directory

[SAFE]: The skill implements a standard and well-structured workflow for repository management using the official GitHub CLI (gh). It follows best practices for issue reporting, such as searching for duplicates and providing clear reproduction steps.
[INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for external data by searching and viewing existing GitHub issues.
Ingestion points: The gh issue list and gh issue view commands retrieve content from external, potentially untrusted sources (issue titles and bodies written by other users).
Boundary markers: The instructions provide logical delimiters and process steps for the agent to evaluate this data against specific "gates."
Capability inventory: The skill can list, view, and create issues using the gh tool.
Sanitization: No explicit sanitization of issue content is mentioned, but the data is used for informational synthesis rather than code execution or prompt construction.

file-issue