The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands like git ls-files, dirname, sort, xargs, and ls to explore the file system and project structure. These are standard tools used for their intended purpose within a development context.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process numerous local project files (e.g., package.json, spec.md, plan.md) to build context. These files constitute an external data source that could contain malicious instructions. The skill lacks explicit boundary markers or sanitization logic to mitigate the risk of the agent following instructions embedded within these files.\n
Ingestion points: Files like CLAUDE.md, product.md, tech-stack.md, tracks.md, and individual track specifications/plans.\n
Boundary markers: None identified in the instructions.\n
Capability inventory: Uses run_shell_command and read_file to access and process project data.\n
Sanitization: No filtering or escaping of content from project files is specified.

context-loader