fix(deploy): mkdir frontend/public/dl before staging the node binary

rsync --delete excludes frontend/public/dl/, so the directory does not exist
on the build host and the `cp` of clawmates-node into it aborted the deploy
(set -e) before anything was pushed.

Co-Authored-By: Claude Opus 4.8 <[email protected]>

crates/cm-api/src/mission_orchestrator.rs

@@ -11,7 +11,13 @@
 //! against the world) is layered on top by Slices 5–8.
 //!
 //! Design notes:
-//!   - Runtime provisioning is opt-in via `RuntimeProvisioner::from_env`.
+//!   - Runtime provisioning is opt-in. Claws are provisioned against the
+//!     mission's OWN daemon (`RuntimeProvisioner::for_gateway` with the
+//!     per-mission endpoint), falling back to the global gateway only when
+//!     there is no per-mission runtime. Provisioning into the global gateway
+//!     while the run executes on a per-mission daemon leaves that daemon
+//!     without the `claw_*` agents — it falls back to the default `scout`
+//!     agent, which cannot see `/mission/repo`.
 //!     Missing runtime = "insert DB rows only, no live claw" — the
 //!     mission still boots; live claws land the moment the runtime
 //!     env is configured + the mission re-launches.
@@ -70,9 +76,13 @@ pub async fn on_launch(
     // running. Falls back silently when docker is unreachable so
     // dev-mode + tests still work — the topology_worker will use the
     // shared runtime endpoint in that case.
+    // The mission's own runtime endpoint. Claws MUST be provisioned against
+    // THIS gateway, not the global one — see RuntimeProvisioner::for_gateway.
+    let mut mission_gateway: Option<String> = None;
     if let Some(prov) = crate::mission_runtime::MissionRuntimeProvisioner::from_env() {
         match prov.ensure_container(mission_id).await {
             Ok(ec) => {
+                mission_gateway = Some(ec.endpoint.clone());
                 let container_name = crate::mission_runtime::container_name(mission_id);
                 if let Err(e) = cm_db::repo::missions::set_runtime_binding(
                     pool,
@@ -157,7 +167,13 @@ pub async fn on_launch(
         );
     }
 
-    let provisioner = RuntimeProvisioner::from_env();
+    // Provision into the mission's own daemon when we have one (so the daemon
+    // that actually runs the turns knows these claws); fall back to the global
+    // gateway only for dev/no-docker setups where the run uses it too.
+    let provisioner = match mission_gateway.clone() {
+        Some(url) => RuntimeProvisioner::for_gateway(url),
+        None => RuntimeProvisioner::from_env(),
+    };
     let mut first_team_id: Option<Uuid> = None;
     let mut provisioned_claws: Vec<cm_domain::AgentId> = Vec::new();
     for (purpose, template_id) in &picks {
@@ -213,17 +229,29 @@ pub async fn on_launch(
     // the freshly-provisioned claws for the run. Non-fatal: without the
     // pin, agents still write (to the sandbox) but the committer can't
     // find the changes in /mission/repo.
-    if !provisioned_claws.is_empty() {
+    if !provisioned_claws.is_empty() && mission_gateway.is_some() {
         if let Some(mp) = crate::mission_runtime::MissionRuntimeProvisioner::from_env() {
-            if let Err(e) = mp
+            match mp
                 .pin_agent_workspaces(mission_id, &provisioned_claws, "/mission/repo")
                 .await
             {
+                Ok(()) => {
+                    // The daemon reads config ONCE at boot and never re-reads
+                    // the file, so the pin is invisible until it restarts. Its
+                    // agents were created through its own config API, so they
+                    // are already persisted to the file and survive the
+                    // restart; the pairing code is re-minted on every launch.
+                    if let Err(e) = mp.restart_container(mission_id).await {
                         eprintln!(
-                    "mission_orchestrator: pin workspaces for mission {mission_id} failed (continuing): {e}"
+                            "mission_orchestrator: restart runtime for {mission_id} failed (continuing, workspace pin will not apply): {e}"
                         );
                     }
                 }
+                Err(e) => eprintln!(
+                    "mission_orchestrator: pin workspaces for mission {mission_id} failed (continuing): {e}"
+                ),
+            }
+        }
     }
 
     // Herdr second-runtime: if runtime_kind='local_herdr', spawn a

crates/cm-api/src/mission_runtime.rs

@@ -482,6 +482,46 @@ fn extract_pairing_code_from_json(body: &str) -> Option<String> {
 }
 
 impl MissionRuntimeProvisioner {
+    /// Restart the mission's runtime container and wait for its gateway to
+    /// answer again. Needed after `pin_agent_workspaces`: the daemon reads
+    /// config once at boot and never re-reads the file, so a file-only setting
+    /// (`workspace.path`, which the config prop API cannot set) only takes
+    /// effect across a restart. Agents provisioned through the daemon's own
+    /// config API are already persisted to that file, so they survive.
+    pub async fn restart_container(&self, mission_id: Uuid) -> Result<(), String> {
+        let name = container_name(mission_id);
+        self.docker
+            .restart_container(&name, None::<bollard::query_parameters::RestartContainerOptions>)
+            .await
+            .map_err(|e| format!("restart mission runtime container: {e}"))?;
+        // Wait for the gateway to serve again so the caller can launch a run
+        // immediately after. ~20s ceiling; the daemon normally boots in ~2s.
+        for _ in 0..40 {
+            tokio::time::sleep(std::time::Duration::from_millis(500)).await;
+            // NOTE: exec_capture only fails on docker errors — a curl that
+            // can't connect still "succeeds" (it prints `curl: (7) …`), so we
+            // must inspect the BODY. /health answers `{"paired":…}`.
+            let body = self
+                .exec_capture(
+                    &name,
+                    vec![
+                        "curl".into(),
+                        "-fsS".into(),
+                        "-m".into(),
+                        "2".into(),
+                        format!("http://127.0.0.1:{GATEWAY_PORT}/health"),
+                    ],
+                )
+                .await
+                .unwrap_or_default();
+            if body.contains("\"paired\"") {
+                eprintln!("mission_runtime: restarted {name}, gateway healthy");
+                return Ok(());
+            }
+        }
+        Err(format!("{name} gateway did not come back after restart"))
+    }
+
     /// Force-remove the mission's runtime container AND its host workspace
     /// dir (the `/mission/repo` checkout bind source). Idempotent: a missing
     /// container or dir is not an error — this is called both by the terminal

crates/cm-api/src/runtime_provision.rs

@@ -66,12 +66,26 @@ impl RuntimeProvisioner {
         let gateway_url = std::env::var("ZEROCLAW_GATEWAY_URL")
             .ok()
             .filter(|u| !u.is_empty())?;
+        Self::for_gateway(gateway_url)
+    }
+
+    /// Build a provisioner aimed at a SPECIFIC gateway, reusing the durable
+    /// `ZEROCLAW_TOKEN`. Mirrors `ZeroClawDriveExecutor::from_env_for_gateway`.
+    ///
+    /// Missions MUST use this with their own per-mission runtime endpoint:
+    /// each mission runs its turns against its own daemon, and that daemon
+    /// loads config once at boot and never re-reads the file. Provisioning a
+    /// mission's claws against the global gateway therefore leaves the
+    /// per-mission daemon with no `claw_*` agents at all — it silently falls
+    /// back to the default agent (`scout`), which is jailed to the global
+    /// workspace and cannot see `/mission/repo`.
+    pub fn for_gateway(gateway_url: String) -> Option<RuntimeProvisioner> {
         let token = std::env::var("ZEROCLAW_TOKEN")
             .ok()
             .filter(|t| !t.is_empty())?;
         Some(RuntimeProvisioner {
             http: reqwest::Client::new(),
-            gateway_url,
+            gateway_url: gateway_url.trim_end_matches('/').to_string(),
             token,
         })
     }

scripts/deploy.sh

@@ -60,6 +60,9 @@ if [ -z "${IMAGES_ONLY:-}" ]; then
   ssh "$BUILD_HOST" 'set -e; cd ~/clawmates
     export PATH=$HOME/.cargo/bin:$PATH CARGO_NET_GIT_FETCH_WITH_CLI=true SQLX_OFFLINE=true
     cargo build --release -p clawmates-node
+    # rsync excludes frontend/public/dl/ (build artifacts), so the dir may not
+    # exist on the build host — create it before staging the daemon binary.
+    mkdir -p frontend/public/dl
     cp target/release/clawmates-node frontend/public/dl/clawmates-node-linux-amd64
     for svc in server frontend; do
       docker build -f images/$svc.Dockerfile -t '"$REGISTRY"'/clawmates/$svc:'"$TAG"' \