Consequence evaluation for AI agents. Check recoverability before destructive actions.
No remote MCP endpoint is published for this listing.
No capability manifest has been published for this listing yet.