The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard command-line utilities including 'rg' (ripgrep) for searching the codebase, and package managers like 'pnpm' and 'npx' for running type checks, linters, and test suites. These commands are constrained to the local project directory and are essential for the skill's core functionality of refactoring and verifying code.- [INDIRECT_PROMPT_INJECTION]: By scanning and ingesting source code from the 'src/' directory, the skill exposes an attack surface for indirect prompt injection where malicious instructions could be embedded in code comments. However, the risk is mitigated by the skill's narrow focus on identifying specific syntactic patterns for refactoring, making it unlikely to be diverted from its primary task.- [DATA_EXPOSURE]: The skill accesses local project-specific configuration and memory files (e.g., '~/.claude/projects/.../memory/feedback_ts_pattern_threshold.md'). This is used to maintain consistency across refactoring tasks within the same project and does not involve exfiltration of sensitive information to external parties.

ts-pattern-refactor