oro-fun-sigidi
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill package is composed entirely of instructional Markdown files and YAML metadata. It contains no executable scripts, shell commands, or binary files.\n- [NO_CODE]: The skill does not define any tools or automated actions. Its purpose is to provide the agent with a conceptual framework for text generation and refactoring tasks within the conversational context.\n- [DATA_EXPOSURE]: No sensitive information, such as environment variables, API keys, or private file paths, is accessed or required by the skill.\n- [REMOTE_CODE_EXECUTION]: There are no external dependencies, package installations, or remote script fetches. All referenced materials are local Markdown files.\n- [PROMPT_INJECTION]: The skill provides instructions for the agent on how to process input text (e.g., 'treat old text as evidence, not authority'). These are functional guidelines for the specified task of refactoring prompts and do not constitute an attempt to subvert the agent's core safety protocols.
Audit Metadata