diff --git a/libpod/container.go b/libpod/container.go
index 4f7fc067e2..2a23952d43 100644
--- a/libpod/container.go
+++ b/libpod/container.go
@@ -236,7 +236,7 @@ type ContainerConfig struct {
 	// UID/GID mappings used by the storage
 	IDMappings storage.IDMappingOptions `json:"idMappingsOptions,omitempty"`
 
-	// Information on the image used for the root filesystem/
+	// Information on the image used for the root filesystem
 	RootfsImageID   string `json:"rootfsImageID,omitempty"`
 	RootfsImageName string `json:"rootfsImageName,omitempty"`
 	// Rootfs to use for the container, this conflicts with RootfsImageID
diff --git a/pkg/specgen/specgen.go b/pkg/specgen/specgen.go
new file mode 100644
index 0000000000..d59e8025b3
--- /dev/null
+++ b/pkg/specgen/specgen.go
@@ -0,0 +1,363 @@
+package specgen
+
+import (
+	"net"
+
+	"github.com/containers/libpod/libpod"
+	spec "github.com/opencontainers/runtime-spec/specs-go"
+)
+
+// TODO
+type Namespace struct {
+}
+
+// ContainerBasicConfig contains the basic parts of a container.
+type ContainerBasicConfig struct {
+	// Name is the name the container will be given.
+	// If no name is provided, one will be randomly generated.
+	// Optional.
+	Name string
+	// Pod is the ID of the pod the container will join.
+	// Optional.
+	Pod string
+	// Entrypoint is the container's entrypoint.
+	// If not given and Image is specified, this will be populated by the
+	// image's configuration.
+	// Optional.
+	Entrypoint []string
+	// Command is the container's command.
+	// If not given and Image is specified, this will be populated by the
+	// image's configuration.
+	// Optional.
+	Command []string
+	// Env is a set of environment variables that will be set in the
+	// container.
+	// Optional.
+	Env map[string]string
+	// Terminal is whether the container will create a PTY.
+	Terminal bool
+	// Stdin is whether the container will keep its STDIN open.
+	Stdin bool
+	// Labels are key-valid labels that are used to add metadata to
+	// containers.
+	// Optional.
+	Labels map[string]string
+	// Annotations are key-value options passed into the container runtime
+	// that can be used to trigger special behavior.
+	// Optional.
+	Annotations map[string]string
+	// StopSignal is the signal that will be used to stop the container.
+	// Must be a non-zero integer below SIGRTMAX.
+	// If not provided, the default, SIGTERM, will be used.
+	// Will conflict with Systemd if Systemd is set to "true" or "always".
+	// Optional.
+	StopSignal *uint
+	// StopTimeout is a timeout between the container's stop signal being
+	// sent and SIGKILL being sent.
+	// If not provided, the default will be used.
+	// If 0 is used, stop signal will not be sent, and SIGKILL will be sent
+	// instead.
+	// Optional.
+	StopTimeout *uint
+	// LogDriver is the container's log driver.
+	// Optional.
+	LogDriver string
+	// LogPath is the path the container's logs will be stored at.
+	// Only available if LogDriver is set to "json-file" or "k8s-file".
+	// Optional.
+	LogPath string
+	// ConmonPidFile is a path at which a PID file for Conmon will be
+	// placed.
+	// If not given, a default location will be used.
+	// Optional.
+	ConmonPidFile string
+	// RestartPolicy is the container's restart policy - an action which
+	// will be taken when the container exits.
+	// If not given, the default policy, which does nothing, will be used.
+	// Optional.
+	RestartPolicy string
+	// RestartRetries is the number of attempts that will be made to restart
+	// the container.
+	// Only available when RestartPolicy is set to "on-failure".
+	// Optional.
+	RestartRetries *uint
+	// OCIRuntime is the name of the OCI runtime that will be used to create
+	// the container.
+	// If not specified, the default will be used.
+	// Optional.
+	OCIRuntime string
+	// Systemd is whether the container will be started in systemd mode.
+	// Valid options are "true", "false", and "always".
+	// "true" enables this mode only if the binary run in the container is
+	// /sbin/init or systemd. "always" unconditionally enables systemd mode.
+	// "false" unconditionally disables systemd mode.
+	// If enabled, mounts and stop signal will be modified.
+	// If set to "always" or set to "true" and conditionally triggered,
+	// conflicts with StopSignal.
+	// If not specified, "false" will be assumed.
+	// Optional.
+	Systemd string
+	// Namespace is the libpod namespace the container will be placed in.
+	// Optional.
+	Namespace string
+
+	// PidNS is the container's PID namespace.
+	// It defaults to private.
+	// Mandatory.
+	PidNS Namespace
+
+	// UtsNS is the container's UTS namespace.
+	// It defaults to private.
+	// Must be set to Private to set Hostname.
+	// Mandatory.
+	UtsNS Namespace
+	// Hostname is the container's hostname. If not set, the hostname will
+	// not be modified (if UtsNS is not private) or will be set to the
+	// container ID (if UtsNS is private).
+	// Conflicts with UtsNS if UtsNS is not set to private.
+	// Optional.
+	Hostname string
+}
+
+// ContainerStorageConfig contains information on the storage configuration of a
+// container.
+type ContainerStorageConfig struct {
+	// Image is the image the container will be based on. The image will be
+	// used as the container's root filesystem, and its environment vars,
+	// volumes, and other configuration will be applied to the container.
+	// Conflicts with Rootfs.
+	// At least one of Image or Rootfs must be specified.
+	Image string
+	// Rootfs is the path to a directory that will be used as the
+	// container's root filesystem. No modification will be made to the
+	// directory, it will be directly mounted into the container as root.
+	// Conflicts with Image.
+	// At least one of Image or Rootfs must be specified.
+	Rootfs string
+	// ImageVolumeMode indicates how image volumes will be created.
+	// Supported modes are "ignore" (do not create), "tmpfs" (create as
+	// tmpfs), and "anonymous" (create as anonymous volumes).
+	// The default is anonymous.
+	// Optional.
+	ImageVolumeMode string
+	// VolumesFrom is a list of containers whose volumes will be added to
+	// this container. Supported mount options may be added after the
+	// container name with a : and include "ro" and "rw".
+	// Optional.
+	VolumesFrom []string
+	// Mounts are mounts that will be added to the container.
+	// These will supersede Image Volumes and VolumesFrom volumes where
+	// there are conflicts.
+	// Optional.
+	Mounts []spec.Mount
+	// Volumes are named volumes that will be added to the container.
+	// These will supersede Image Volumes and VolumesFrom volumes where
+	// there are conflicts.
+	// Optional.
+	Volumes []*libpod.ContainerNamedVolume
+	// Devices are devices that will be added to the container.
+	// Optional.
+	Devices []spec.LinuxDevice
+	// IpcNS is the container's IPC namespace.
+	// Default is private.
+	// Conflicts with ShmSize if not set to private.
+	// Mandatory.
+	IpcNS Namespace
+	// ShmSize is the size of the tmpfs to mount in at /dev/shm, in bytes.
+	// Conflicts with ShmSize if ShmSize is not private.
+	// Optional.
+	ShmSize *int64
+	// WorkDir is the container's working directory.
+	// If unset, the default, /, will be used.
+	// Optional.
+	WorkDir string
+	// RootfsPropagation is the rootfs propagation mode for the container.
+	// If not set, the default of rslave will be used.
+	// Optional.
+	RootfsPropagation string
+}
+
+// ContainerSecurityConfig is a container's security features, including
+// SELinux, Apparmor, and Seccomp.
+type ContainerSecurityConfig struct {
+	// Privileged is whether the container is privileged.
+	// Privileged does the following:
+	// - Adds all devices on the system to the container.
+	// - Adds all capabilities to the container.
+	// - Disables Seccomp, SELinux, and Apparmor confinement.
+	// TODO: this conflicts with things.
+	// TODO: this does more.
+	Privileged bool
+	// User is the user the container will be run as.
+	// Can be given as a UID or a username; if a username, it will be
+	// resolved within the container, using the container's /etc/passwd.
+	// If unset, the container will be run as root.
+	// Optional.
+	User string
+	// Groups are a list of supplemental groups the container's user will
+	// be granted access to.
+	// Optional.
+	Groups []string
+	// CapAdd are capabilities which will be added to the container.
+	// Conflicts with Privileged.
+	// Optional.
+	CapAdd []string
+	// CapDrop are capabilities which will be removed from the container.
+	// Conflicts with Privileged.
+	// Optional.
+	CapDrop []string
+	// SelinuxProcessLabel is the process label the container will use.
+	// If SELinux is enabled and this is not specified, a label will be
+	// automatically generated if not specified.
+	// Optional.
+	SelinuxProcessLabel string
+	// SelinuxMountLabel is the mount label the container will use.
+	// If SELinux is enabled and this is not specified, a label will be
+	// automatically generated if not specified.
+	// Optional.
+	SelinuxMountLabel string
+	// SelinuxOpts are options for configuring SELinux.
+	// Optional.
+	SelinuxOpts []string
+	// ApparmorProfile is the name of the Apparmor profile the container
+	// will use.
+	// Optional.
+	ApparmorProfile string
+	// SeccompProfilePath is the path to a JSON file containing the
+	// container's Seccomp profile.
+	// If not specified, no Seccomp profile will be used.
+	// Optional.
+	SeccompProfilePath string
+	// NoNewPrivileges is whether the container will set the no new
+	// privileges flag on create, which disables gaining additional
+	// privileges (e.g. via setuid) in the container.
+	NoNewPrivileges bool
+	// UserNS is the container's user namespace.
+	// It defaults to host, indicating that no user namespace will be
+	// created.
+	// If set to private, IDMappings must be set.
+	// Mandatory.
+	UserNS Namespace
+	// IDMappings are UID and GID mappings that will be used by user
+	// namespaces.
+	// Required if UserNS is private.
+	IDMappings storage.IDMappingOptions
+}
+
+// ContainerCgroupConfig contains configuration information about a container's
+// cgroups.
+type ContainerCgroupConfig struct {
+	// CgroupNS is the container's cgroup namespace.
+	// It defaults to private.
+	// Conflicts with NoCgroups if not set to host.
+	// Mandatory.
+	CgroupNS Namespace
+	// NoCgroups indicates that the container should not create CGroups.
+	// Conflicts with CgroupParent and CgroupNS if CgroupNS is not set to
+	// host.
+	NoCgroups bool
+	// CgroupParent is the container's CGroup parent.
+	// If not set, the default for the current cgroup driver will be used.
+	// Conflicts with NoCgroups.
+	// Optional.
+	CgroupParent string
+}
+
+// ContainerNetworkConfig contains information on a container's network
+// configuration.
+type ContainerNetworkConfig struct {
+	// NetNS is the configuration to use for the container's network
+	// namespace.
+	// Mandatory.
+	NetNS Namespace
+	// ConfigureNetNS is whether Libpod will configure the container's
+	// network namespace to send and receive traffic.
+	// Only available is NetNS is private - conflicts with other NetNS
+	// modes.
+	ConfigureNetNS bool
+	// StaticIP is the a IPv4 address of the container.
+	// Only available if ConfigureNetNS is true.
+	// Optional.
+	StaticIP *net.IP
+	// StaticIPv6 is a static IPv6 address to set in the container.
+	// Only available if ConfigureNetNS is true.
+	// Optional.
+	StaticIPv6 *net.IP
+	// StaticMAC is a static MAC address to set in the container.
+	// Only available if ConfigureNetNS is true.
+	// Optional.
+	StaticMAC *net.HardwareAddr
+	// PortBindings is a set of ports to map into the container.
+	// Only available if ConfigureNetNS is true.
+	// Optional.
+	PortMappings []ocicni.PortMapping
+	// PublishImagePorts will publish ports specified in the image to random
+	// ports outside.
+	// Requires Image to be set.
+	PublishImagePorts bool
+	// CNINetworks is a list of CNI networks to join the container to.
+	// If this list is empty, the default CNI network will be joined
+	// instead. If at least one entry is present, we will not join the
+	// default network (unless it is part of this list).
+	// Only available if ConfigureNetNS is true.
+	// Optional.
+	CNINetworks []string
+	// UseImageResolvConf indicates that resolv.conf should not be managed
+	// by Podman, but instead sourced from the image.
+	// Conflicts with DNSServer, DNSSearch, DNSOption.
+	UseImageResolvConf bool
+	// DNSServer is a set of DNS servers that will be used in the
+	// container's resolv.conf, replacing the host's DNS Servers which are
+	// used by default.
+	// Conflicts with UseImageResolvConf.
+	// Optional.
+	DNSServer []net.IP
+	// DNSSearch is a set of DNS search domains that will be used in the
+	// container's resolv.conf, replacing the host's DNS search domains
+	// which are used by default.
+	// Conflicts with UseImageResolvConf.
+	// Optional.
+	DNSSearch []string
+	// DNSOption is a set of DNS options that will be used in the
+	// container's resolv.conf, replacing the host's DNS options which are
+	// used by default.
+	// Conflicts with UseImageResolvConf.
+	// Optional.
+	DNSOption []string
+	// UseImageHosts indicates that /etc/hosts should not be managed by
+	// Podman, and instead sourced from the image.
+	// Conflicts with HostAdd.
+	UseImageHosts bool
+	// HostAdd is a set of hosts which will be added to the container's
+	// /etc/hosts file.
+	// Conflicts with UseImageHosts.
+	// Optional.
+	HostAdd []string
+}
+
+// ContainerResourceConfig contains information on container resource limits.
+type ContainerResourceConfig struct {
+	// ResourceLimits are resource limits to apply to the container.
+	// Can only be set as root on cgroups v1 systems, but can be set as
+	// rootless as well for cgroups v2.
+	// Optional.
+	ResourceLimits *spec.LinuxResources
+	// Rlimits are POSIX rlimits to apply to the container.
+	// Optional.
+	Rlimits []spec.POSIXRlimit
+	// OOMScoreAdj adjusts the score used by the OOM killer to determine
+	// processes to kill for the container's process.
+	// Optional.
+	OOMScoreAdj *int
+}
+
+// SpecGenerator creates an OCI spec and Libpod configuration options to create
+// a container based on the given configuration.
+type SpecGenerator struct {
+	ContainerBasicConfig
+	ContainerStorageConfig
+	ContainerSecurityConfig
+	ContainerCgroupConfig
+	ContainerNetworkConfig
+	ContainerResourceConfig
+}