The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands via the claude CLI to automate development workflows. This includes using the -p flag for one-shot tasks and managing interactive sessions, which is the primary intended function of the tool.
[EXTERNAL_DOWNLOADS]: The documentation references the installation of the @anthropic-ai/claude-code package from the official npm registry. This is a reference to an official tool from a well-known service.
[PROMPT_INJECTION]: The skill allows the agent to process project-specific files like CLAUDE.md and source code, which constitutes an indirect prompt injection surface. \n
Ingestion points: Project files, git diffs, and configuration files (e.g., CLAUDE.md) read by the CLI. \n
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are provided in the skill text. \n
Capability inventory: The skill uses a bash tool to run commands that can modify the file system and execute further shell instructions. \n
Sanitization: No specific sanitization or filtering of input data is mentioned.

claude-code