onboarding
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of conversational instructions for an AI agent to interview a user about their business goals and brand personality. It does not include any executable scripts, binary files, or external dependencies.
- [NO_CODE]: There are no programming language implementations (Python, Node.js, Shell, etc.) within the skill. Its primary function is to generate a structured markdown file (
SITE.md) based on user responses. - [DATA_EXPOSURE_EXFILTRATION]: The skill asks for business-related information (name, goals, audience), which is appropriate for its stated purpose of project onboarding. It does not attempt to access sensitive system files, environment variables, or hardcoded credentials.
- [COMMAND_EXECUTION]: There are no instructions that would lead the agent to execute shell commands or system-level operations.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided text to populate
SITE.md, the interaction is limited to text generation within a markdown file. There are no high-privilege tools or network-connected capabilities associated with this skill that could be abused through malicious user input.
Audit Metadata