The Agent Skills Directory

[COMMAND_EXECUTION]: The skill initiates a local prerequisite check by running .specify/scripts/powershell/check-prerequisites.ps1 and utilizes standard git commands (git rev-parse) to verify repository status.
[PROMPT_INJECTION]: The skill processes and executes instructions found in local project files, creating a vulnerability surface for indirect prompt injection.
Ingestion points: tasks.md, plan.md, data-model.md, and checklists in the .specify/ or feature directory.
Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands within the ingested files.
Capability inventory: The skill allows the agent to modify the file system (e.g., creating .gitignore, .dockerignore) and execute arbitrary shell commands required by the implementation tasks.
Sanitization: The skill provides specific instructions to escape single quotes in user arguments ($ARGUMENTS) to prevent command injection at the prerequisite check step.

speckit-implement