The Agent Skills Directory

[SAFE]: The skill acts as a legitimate orchestration tool for story auditing and exhibits no malicious behavior or unauthorized data access.- [COMMAND_EXECUTION]: The skill executes local Node.js scripts provided in the package (such as 'check-ai-patterns.js' and 'normalize-punctuation.js') to perform static analysis on text files. These scripts utilize regular expressions to identify stylistic issues and do not perform network or privileged operations.- [REMOTE_CODE_EXECUTION]: The skill uses the platform's sub-agent spawning mechanism to delegate specialized analysis tasks to other project-local agents. This is a standard orchestration feature and occurs within the project's execution boundaries.- [PROMPT_INJECTION]: The instructions explicitly define the agent's role as finding problems within the provided text, which encourages a defensive posture that treats untrusted input as data rather than instructions.- [EXTERNAL_DOWNLOADS]: The skill's metadata contains a URL to the author's GitHub repository for source tracking, which is a recognized and trusted hosting service.

story-review