launchdarkly-automations-archival-settings
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements strong safety measures for administrative actions. It explicitly instructs the agent to 'Confirm before every write' and enforces a 'human-in-the-loop' policy for enabling archival, ensuring that the AI cannot bypass production safety checks on its own.
- [SAFE]: Data access is restricted to official LaunchDarkly MCP tools (
get-automations,update-automations). The skill does not attempt to make direct REST API calls or access unauthorized external endpoints. - [SAFE]: Dependencies are limited to '@launchdarkly/mcp-server', which is a legitimate vendor resource consistent with the skill's purpose and author.
- [SAFE]: The fallback instructions in 'references/browser-steps.md' are safe, guiding the user to standard LaunchDarkly dashboard URLs for manual configuration when tools are unavailable.
Audit Metadata