command-center-builder

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill functions as a guided configuration assistant and does not possess capabilities to perform network requests, access the filesystem, or execute commands.
  • [NO_CODE]: Analysis confirms the skill is composed strictly of Markdown instructions and templates; no Python, JavaScript, or shell scripts are present in the package.
  • [PROMPT_INJECTION]: The skill's instructions and persona definitions (e.g., JARVIS, ALFRED) are designed to tailor the user experience and do not include patterns to override safety guardrails or perform prompt extraction.
  • [DATA_EXFILTRATION]: While the skill requests user writing samples for style analysis, it lacks any mechanism for external data transmission or persistent storage, ensuring information remains within the conversation context.
  • [EXTERNAL_DOWNLOADS]: The documentation mentions various third-party integrations (Google, Slack, Atlassian) as setup options for the user's project, but the skill itself does not download or execute any external resources.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 03:29 AM