The Agent Skills Directory

[COMMAND_EXECUTION]: The skill implements a PreToolUse hook that executes a local Python script dev-delegation-guard.py using uv run to manage delegation permissions.
[COMMAND_EXECUTION]: Utilizes dynamic context injection with the !cat`` command to pull local reference documents into the prompt at load time. This is used for including development constraints.
[REMOTE_CODE_EXECUTION]: Employs dynamic loading to transition between different workflow phases by reading relative paths to other skills (e.g., ../../skills/dev-explore/SKILL.md). This is a local modularity pattern rather than remote execution.
[PROMPT_INJECTION]: There is a surface for indirect prompt injection as the skill reads workflow state and requirements from local files like .planning/HANDOFF.md and .planning/SPEC.md. While these are processed without explicit boundary markers, they are part of the intended state management for a development environment.

dev