The Agent Skills Directory

[COMMAND_EXECUTION]: The 'Setup' section of the skill includes a shell command used to manage the project's .gitignore file. It ensures the .skills/ directory is ignored and commits this change to the repository. This is a standard project initialization task carried out in the local development environment.
[PROMPT_INJECTION]: The skill reads and processes instructions from external project files, which creates a surface for indirect prompt injection.
Ingestion points: The skill retrieves instructions from tasks.md, .skills/progress.md, and docs/agents/project.md as part of its operational loop.
Boundary markers: The skill does not employ specific delimiters or boundary markers for the data ingested from these files.
Capability inventory: The skill possesses capabilities to write files, manage git state, and execute tests via the tdd sub-skill.
Sanitization: Content from these task files is utilized as instruction sets without a validation or sanitization phase. This behavior is inherent to the skill's primary function of executing tasks defined in a project's documentation.

execute-inline