configs-update
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references a requirement for a remotely hosted LaunchDarkly Model Context Protocol (MCP) server. This is documented neutrally as a standard operational requirement for the skill's stated purpose of managing LaunchDarkly configurations.
- [SAFE]: No security concerns such as prompt injection, data exfiltration, obfuscation, or unauthorized persistence mechanisms were identified. The workflow incorporates safety measures, including health checks (
get-ai-config-health) and a preference for reversible archival over permanent deletion. - [SAFE]: The skill instructs the agent to obtain explicit user confirmation before performing irreversible actions, such as deleting configurations or variations.
Audit Metadata