The Agent Skills Directory

[SAFE]: The skill is strictly instructional and focuses on improving code quality through established refactoring patterns. No malicious instructions or hidden behavior were identified.\n- [COMMAND_EXECUTION]: The skill mentions the use of development tools like npm, npx, and tsc for verifying code changes. These commands are localized to the development environment and are standard for the described tasks.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided source code, which serves as an untrusted input surface. Malicious instructions could be embedded in code comments, though this is a baseline risk for all code-processing skills. The skill relies on standard linting and testing for validation.\n
Ingestion points: User-provided source code processed during refactoring steps (SKILL.md).\n
Boundary markers: No specific boundary markers or 'ignore' instructions are provided for the agent to separate code from embedded comments.\n
Capability inventory: Use of shell execution via npm and npx to run tests and linters (Step B and Workflow Example).\n
Sanitization: No input sanitization is described for the code being processed.

code-refactoring