The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external, potentially untrusted sources such as GitHub PR discussions, linked tickets, and local files like TODO.md or PLAN.md.
Ingestion points: Content is gathered during the 'Establish Scope And Authority' and 'Follow CI And Review To Convergence' phases.
Boundary markers: There are no explicit instructions or delimiters defined to separate agent instructions from the content being read from external files or PR threads.
Capability inventory: The skill has broad capabilities including local file system modification (fixing code), git operations (committing, pushing, and merging), and PR management via the GitHub CLI or API.
Sanitization: The instructions do not mention any sanitization, validation, or escaping of the content retrieved from external sources before it is processed by the LLM.
[COMMAND_EXECUTION]: The skill relies on executing various shell-based tools and Git commands to manage the repository state.
Evidence: It explicitly orchestrates git push, gh-stack, and other specialized Git/GitHub management skills to modify the remote state of the repository.

git-finish