The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local test scripts (e.g., npm test, npm run test:integration) to validate refactored code. While essential for the skill's purpose, this allows for the execution of arbitrary commands defined in the project's configuration.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests and processes untrusted code files for refactoring. Ingestion points: Reads all files from the 'task Implementation Record' and project configuration files. Boundary markers: No explicit delimiters or instructions are used to isolate untrusted code content from the agent's instructions. Capability inventory: The skill can modify local files and execute shell commands. Sanitization: No sanitization or filtering is applied to the code input before analysis.

refactor-code