The Agent Skills Directory

[COMMAND_EXECUTION]: The skill interacts with the local system by executing several Git commands to manage the repository state during orchestration.
Evidence: It runs git status --short, git add -A (after user confirmation and after each ticket), and instructs subagents to run git add -N . to track new files without staging them.
[INDIRECT_PROMPT_INJECTION]: The skill processes external ticket files and passes their contents to subagents as primary implementation instructions, creating a vulnerability surface where untrusted data could influence agent behavior.
Ingestion points: Ticket files (Markdown or text) stored in a specified directory or provided as a list.
Boundary markers: The skill does not implement explicit delimiters or instruction-isolation markers when passing the ticket path/content to subagents.
Capability inventory: The orchestrator has the ability to execute shell commands (git), and the subagents it creates are explicitly described as having "full tool access."
Sanitization: There is no logic provided to sanitize or validate the contents of the ticket files before they are utilized by the subagents.

delegate-tickets