The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by fetching and processing external data from GitHub issues.
Ingestion points: Issue bodies are fetched via gh issue view in references/clarify.md and issue comments are retrieved in references/wrap-up.md.
Boundary markers: The instructions lack explicit delimiters or instructions to ignore embedded commands when processing fetched content.
Capability inventory: The skill possesses significant capabilities including Bash(gh:*), Bash(git:*), Write, and Edit as defined in SKILL.md.
Sanitization: There is no explicit sanitization or filtering of the ingested GitHub content before it is interpolated into prompts or used to generate plans.
[COMMAND_EXECUTION]: The skill routinely executes CLI commands to interact with the repository and GitHub project.
Evidence: Uses gh and git commands throughout all reference files for branch management, issue updates, and pull request creation.

ghpm-work