The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from GitHub issues (titles, bodies, and comments) and uses it to drive further actions such as creating sub-issues or updating project fields. A malicious issue description could attempt to influence the agent's behavior. * Ingestion points: Issue metadata and body content are fetched via gh issue view and gh api in SKILL.md during UPDATE, QUERY, and FIND modes. * Boundary markers: The instructions do not define explicit boundary markers or delimiters for data ingested from GitHub. * Capability inventory: The skill uses gh issue create, gh issue edit, and various gh api mutations to modify resources in the dotCMS/core repository. * Sanitization: No specific sanitization or escaping logic is mandated for external content before it is interpolated into commands or new issue bodies.
[COMMAND_EXECUTION]: The skill uses several shell commands to interact with the environment and GitHub. * It performs repository checks using git branch --show-current. * It manages local state by reading from and writing to ~/.config/dotcms/create-issue/default-team using mkdir, cat, and echo. * It uses the GitHub CLI (gh) extensively for API interactions and authentication refreshing.

dot-issue-manage