favicon-generator
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill does not contain any instructions aimed at overriding system prompts, bypassing safety filters, or extracting internal instructions.
- [DATA_EXPOSURE]: There are no hardcoded credentials, API keys, or unauthorized access to sensitive system files. The skill references brand context files such as
.claude/product-marketing-context.mdto align with the user's project identity, which is standard and safe behavior for agent-based development tools. - [REMOTE_CODE_EXECUTION]: No remote code execution patterns, such as piping network downloads to a shell, were detected. The mention of the 'favicons' npm package is a tool recommendation for the user and not an automated execution script.
- [COMMAND_EXECUTION]: The skill does not invoke any system commands, subprocesses, or shell scripts.
- [DATA_EXFILTRATION]: No network operations or attempts to transmit data to external domains were found.
Audit Metadata