The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it reads and processes data from untrusted repository contents and metadata.
Ingestion points: Reads all markdown files in the repository (Step 2), git commit logs (Step 1), and Pull Request body content (Step 8).
Boundary markers: Absent. There are no instructions or delimiters defined to distinguish between user instructions and documentation data, nor are there warnings to ignore embedded instructions.
Capability inventory: The skill has file-write access, and can execute git commit, git push, and gh pr edit.
Sanitization: No sanitization or validation is performed on the ingested documentation content or PR body before the agent processes it.
[COMMAND_EXECUTION]: The skill executes shell commands to perform repository management and automation.
Evidence: Uses git diff, git log, git commit, git push, gh pr view, and gh pr edit to synchronize files and update the remote repository.

document-update