The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes internal Python scripts and standard system utilities like git to perform its primary functions. For instance, verify_scope.py executes git diff to identify modified files, and test_verify_scope.py uses git init for setting up test environments. These actions are standard for developer tools and are used here to enforce security boundaries.
[INDIRECT_PROMPT_INJECTION]: The skill implements a workflow where an allowlist is generated from a refactoring specification file, creating a surface for processing untrusted data.
Ingestion points: The scripts/generate_allowlist.py script ingests the refactoring specification file provided by the user.
Boundary markers: The script searches for specific Markdown headers (IN SCOPE, OUT OF SCOPE) and checkbox patterns (- [x]) to identify relevant instructions.
Capability inventory: The skill's primary capabilities include codebase-wide file modification and execution of git-based verification commands.
Sanitization: The extraction logic focuses on specific character patterns for paths and globs while ignoring general prose, which serves to sanitize the input and prevent the accidental execution of instructions embedded in the spec.

large-scale-refactor