The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code as its primary input, which could contain malicious instructions designed to influence the agent.
Ingestion points: Code snippets and project files accessed via Read and Grep tools.
Boundary markers: Absent; there are no instructions to use delimiters or ignore instructions within the code being refactored.
Capability inventory: The skill is permitted to use Write and Edit tools to modify files on the system.
Sanitization: No explicit instructions for sanitizing or validating the input code are provided.
[COMMAND_EXECUTION]: The skill suggests using shell commands to verify code integrity and behavior preservation.
Evidence: Shell commands like npm test, npx tsc, and codex-cli shell are documented for use in the validation phase.
Context: These commands are typical for development environments and are used here for legitimate verification purposes.

code-refactoring