launchdarkly-release-autopilot-onboarding
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements robust safety protocols, including a mandatory 'Gather, confirm, then write' workflow that prevents any modifications until the user explicitly approves an itemized plan.\n- [SAFE]: Access to sensitive data is handled with care. The user's email address is retrieved to provide specific guidance for browser-based authentication steps, ensuring the user is signed into the correct account, rather than being exfiltrated to external systems.\n- [SAFE]: The skill operates within a trusted vendor context, utilizing LaunchDarkly's official MCP server and linking to established domains (
launchdarkly.com) and repositories (github.com/launchdarkly).\n- [SAFE]: It employs the principle of least privilege by recommending configurations that require human approval for automated actions (like auto-flagging) and making more autonomous features opt-in.
Audit Metadata