The Agent Skills Directory

[SAFE]: The skill leverages standard developer utilities (git, gh) to perform routine code and pull request management tasks. It advocates for security best practices, such as treating security and privacy findings from CI as blocking issues.
[COMMAND_EXECUTION]: The skill utilizes command-line tools including git, gh (GitHub CLI), and the vendor-specific wendy CLI (e.g., wendy deploy) to interact with the repository and deployment pipelines.
[INDIRECT_PROMPT_INJECTION]: The skill has a functional requirement to ingest and process data from external and potentially untrusted sources, including issue titles from trackers (e.g., Linear), PR comments, human reviews, and CI feedback.
Ingestion points: Processes content from external issue links, GitHub PR descriptions, comments, and automated review findings as outlined in SKILL.md and references/descriptions.md.
Boundary markers: No specific delimiters or instructions to ignore adversarial content within these inputs are defined.
Capability inventory: The skill uses gh pr create and gh pr edit to modify pull request content based on gathered information.
Sanitization: No explicit sanitization or validation protocols for the external text are described in the instructions.

pull-request