The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data from GitHub issues to drive agent behavior, creating a vulnerability surface for indirect prompt injection.
Ingestion points: The skill reads the current status and content of issues from GitHub projects and repositories (SKILL.md).
Boundary markers: The instructions do not define delimiters or provide warnings to ignore instructions embedded within external issue descriptions or comments.
Capability inventory: The skill can dispatch sub-agents, create work trees (file system modification), create and comment on GitHub issues (network operations), and modify local task files (SKILL.md).
Sanitization: There is no mention of sanitizing issue content or validating GitHub labels and statuses before processing them as instructions for state transitions.
[COMMAND_EXECUTION]: The skill instructs the agent to "create new work trees" for confirmed issues, which involves executing git worktree commands on the local system (SKILL.md).

agile-proces