docs(skill): align release-ready host guidance

skills/apsuflow/SKILL.md

@@ -46,11 +46,15 @@ in the same change when applicable requirements or CLI behavior change.
   profiles contain only host configuration and service scripts. The daemon never
   detects a distribution, libc, or init system. Before opening its listener it
   must prove its unprivileged effective UID and create/remove a child beneath a
-  writable delegated cgroup v2 path exposing `cpu`, `memory`, and `pids`.
+  writable delegated cgroup v2 parent exposing `cpu`, `memory`, and `pids` while
+  the daemon itself remains in a leaf cgroup. On systemd profiles this requires
+  `Delegate=yes` plus `ProtectControlGroups=private`; Apsuflow then moves itself
+  into `/daemon` before proving or creating workload children.
   `doctor` reports the same kernel-observed parent-process and cgroup facts.
-  Failure is closed: there is no root or service-manager-specific fallback. The
-  current runit profile also passed its complete acceptance path on Void Linux
-  musl; that is evidence for the capability contract, not a distro special case.
+  Failure is closed: there is no root or service-manager-specific fallback.
+  Both release-supported profiles have complete acceptance proof, including the
+  Void-musl/runit PID-1 path in a local KVM/QEMU VM with a real OCI workload,
+  restart, limits, and encrypted backup/restore.
 - `.apsu` contains desired workloads and infrastructure, never secret values or
   mutable runtime state.
 - Services are continuously reconciled. Jobs are finite run-to-completion work.
@@ -76,15 +80,22 @@ that drift before making another change.
 ## Install and update
 
 Use only the exact stable tag and artifact named by `docs/getting-started.md`.
-Verify `SHA256SUMS` and its SSH signature before installing; never use a mutable
-`latest` URL. Keep the previous executable and state backup until the new one
+Official bootstrap and Cloud-Init entry points require that exact version,
+verify `SHA256SUMS` with the repository-pinned SSH release identity, verify the
+selected archive hash, replace atomically, and retain `apsuflow.previous`; never
+use a mutable `latest` URL. Keep the previous executable and state backup until the new one
 passes `version`, `doctor`, workload diagnosis, and each operator-facing route.
-For a private image registry, pipe the credential into `apsuflow registry login
-<host> --username <user> --password-stdin`; do not put it in `.apsu` or shell
-history. Login stores sealed `_registry/{host}` state, and joined agents receive
-only their per-node-resealed pull credential during assignment. Confirm the next
-pull succeeds; `unauthorized` means the registry host or credential is wrong, not
-that the workload should be stopped first.
+For a private image registry, pipe the exact registry credential in its
+`username:password` form into `apsuflow registry login <host> --password-stdin`;
+the command has no separate username flag. Do not put the credential in `.apsu`,
+process arguments, shell history, a runtime auth file, or host-side container-tool
+configuration. Login stores sealed `_registry/{host}` state, and joined agents
+receive only their per-node-resealed pull credential during assignment. Establish
+and verify this login before the first apply, then confirm the digest-pinned pull
+and workload health through Apsuflow. If a known-good credential still yields
+`unauthorized`, stop: that is an Apsuflow credential-delivery defect, not permission
+to use `skopeo login`, copy an OCI archive, invoke the runtime, or otherwise bypass
+the control plane.
 
 For host-network workloads, confirm the old container released its host ports and
 the successor is the sole running generation; repeated replacement failures are
@@ -112,10 +123,12 @@ host paths—also require `apply --acknowledge-recovery-risk`; read the reported
 source location and consequence before accepting it. Verify the actual route or
 job result; a successful apply alone proves only admission.
 
-The daemon normally needs root for host effects. Do not therefore run every
-client command with `sudo`: protect the daemon data directory and use an operator
-context or token. Never make bootstrap tokens or the data directory broadly
-readable. `server --no-auth` is acceptable only for disposable loopback use.
+The server daemon runs unprivileged in its delegated cgroup; the agent remains
+rootful only for its authorized runtime and host-network effects. Do not run
+client commands with `sudo`: protect daemon state and use an operator context or
+token. Join credentials are node-bound, role-bound, single-use, and expire after
+15 minutes; never reuse, commit, or broaden them. `server --no-auth` is
+acceptable only for disposable loopback use.
 
 ## Multiple environments
 
@@ -188,11 +201,13 @@ job backup {
 
 ## Cluster and network
 
-Start the first server on a stable address. Create a short-lived role-scoped join
-token with `apsuflow token create --name NODE --role server`, then start another
-server with `apsuflow server --peer FIRST:7946 --token TOKEN`. For an agent,
-create an `agent` token and run `apsuflow agent --join FIRST:7654 --token TOKEN`.
-Never reuse or commit a bootstrap token.
+Start the first server on a stable address. Create a 15-minute, single-use join
+credential bound to the exact node ID with `apsuflow token create --name NODE
+--role server`, then start that server with `apsuflow server --peer FIRST:7946
+--token TOKEN`. For an agent, mint the same node-bound credential with `--role
+agent` and run `apsuflow agent --node-id NODE --join FIRST:7654 --token TOKEN`.
+The `--node-id` value must exactly match the credential name. Never reuse or
+commit a join credential.
 
 Allow between nodes only required control-plane traffic:
 
@@ -241,6 +256,10 @@ apsuflow status --context recovery
 
 This backs up Apsuflow control-plane state, not bind-mounted application data,
 databases, object stores, or brokers. Back those up natively and test both paths.
+Restore replaces authentication state and removes the empty target's bootstrap
+token because it cannot match restored credential hashes. Retain an original
+administrator credential outside the lost server data directory; without one,
+recovery is blocked rather than silently granting new access.
 
 Never mutate `store.db`, WAL files, generations, assignments, certificates, or
 sealed values with SQLite/SQL. Never invoke crun/youki or host-network tools to