fixes to operator restore flow

.context/TASKS.md

@@ -35,9 +35,9 @@ the name-based policy work.
 -->
 
 ### Phase 1: Correctness & Broken Things `#priority:high`
-- [ ] Fix broken CI integration test #source:jira.xml #added:2026-07-14
-- [ ] Fix broken recovery/restore flow #source:jira.xml #added:2026-07-14
-- [ ] Fix `spike cipher` stream mode (broken; owner: Murat); JSON mode fix unblocks encryption-as-a-service demo/docs #source:jira.xml #added:2026-07-14
+- [x] Fix broken CI integration test #source:jira.xml #added:2026-07-14 #done:2026-07-16 (stale: CI has been green for several weeks; closed on user confirmation)
+- [ ] Fix broken recovery/restore flow #source:jira.xml #added:2026-07-14 (code review 2026-07-16: no live breakage found; awaiting the scripted drill in Phase 3 to confirm and close)
+- [x] Fix `spike cipher` stream mode (broken; owner: Murat); JSON mode fix unblocks encryption-as-a-service demo/docs #source:jira.xml #added:2026-07-14 #done:2026-07-15 (stale: both cipher streaming and file modes verified passing via the make start checks on 2026-07-15)
 - [ ] Retry sqlite operations with exponential backoff on transient locks (all of `app/nexus/internal/state/persist`) → ideas/research-db-resilience.md #source:jira.xml #added:2026-07-14
 - [ ] Bound the Bootstrap keeper-wait loop with a configurable timeout/max-attempts instead of looping forever → ideas/research-db-resilience.md #source:jira.xml #added:2026-07-14
 - [ ] Make `env` accessors return sentinel errors instead of calling `log.FatalLn` (removes env→log circular dep, makes them testable) → ideas/research-env-error-handling.md #source:jira.xml #added:2026-07-14
@@ -50,6 +50,7 @@ the name-based policy work.
 - [ ] Add integration tests: root key cached/recovered/not-re-initialized; secret & policy CRUD; Pilot denies when Nexus uninitialized / warns when unreachable → ideas/research-cli-testing.md #source:jira.xml #added:2026-07-14
 - [ ] Raise CLI command coverage to 60%+ via unit + HTTP-mock tests; fix `t.Skip()`ed tests; DI-refactor `sendShardsToKeepers` → ideas/research-cli-testing.md #source:jira.xml #added:2026-07-14
 - [ ] `start.sh` should exercise recovery/restore and encryption/decryption #source:jira.xml #added:2026-07-14
+- [ ] Scripted live recovery/restore drill: once `make start` completes cleanly, run `spike operator recover`, kill Nexus and the Keepers, restart Nexus alone, feed the shards back via `spike operator restore` (scriptable via stdin since fix/operator-restore), and verify a pre-crash secret reads back. Rationale: the 2026-07-16 code review found no live breakage (shard-index fidelity intact end to end; guards use exact SPIFFE role matching, unaffected by the policy-name migration), so only a drill can prove the Phase 1 "recovery/restore is broken" claim stale and close both tasks. Needs the recover/restore role entries (spire-server-entry-recover-register.sh / -restore-register.sh), which make start does not register by default. #added:2026-07-16
 
 ### Phase 4: Policy & Secrets `#priority:medium`
 - [ ] Add a `list` permission type; scope `spike secret list` to the caller's allowed path patterns → ideas/research-list-permission.md #source:jira.xml #added:2026-07-14

app/nexus/internal/route/operator/recover.go

@@ -12,6 +12,7 @@ import (
 	"github.com/spiffe/spike-sdk-go/config/env"
 	sdkErrors "github.com/spiffe/spike-sdk-go/errors"
 	"github.com/spiffe/spike-sdk-go/journal"
+	"github.com/spiffe/spike-sdk-go/log"
 	"github.com/spiffe/spike-sdk-go/net"
 	"github.com/spiffe/spike-sdk-go/security/mem"
 
@@ -48,6 +49,22 @@ func RouteRecover(
 	const fName = "routeRecover"
 	journal.AuditRequest(fName, r, audit, journal.AuditCreate)
 
+	// The in-memory backend keeps no persistent state, so recovery
+	// shards would be useless after a restart. Reject the request
+	// explicitly instead of failing later with a misleading "not
+	// enough shards" internal error.
+	if env.BackendStoreTypeVal() == env.Memory {
+		log.Warn(fName, "message", "rejecting recover: in-memory backend")
+		failErr := sdkErrors.ErrDataInvalidInput.Clone()
+		failErr.Msg = "recovery is not applicable to the in-memory backend"
+		if respondErr := net.Fail(
+			reqres.RecoverResponse{}.BadRequest(), w, http.StatusBadRequest,
+		); respondErr != nil {
+			return failErr.Wrap(respondErr)
+		}
+		return failErr
+	}
+
 	_, err := net.ReadParseAndGuard[
 		reqres.RecoverRequest, reqres.RecoverResponse](
 		w, r, reqres.RecoverResponse{}.BadRequest(), guardRecoverRequest,

app/nexus/internal/route/operator/recover_test.go

@@ -0,0 +1,59 @@
+//    \\ SPIKE: Secure your secrets with SPIFFE. — https://spike.ist/
+//  \\\\\ Copyright 2024-present SPIKE contributors.
+// \\\\\\\ SPDX-License-Identifier: Apache-2.0
+
+package operator
+
+import (
+	"net/http"
+	"net/http/httptest"
+	"os"
+	"testing"
+
+	"github.com/spiffe/spike-sdk-go/config/env"
+	sdkErrors "github.com/spiffe/spike-sdk-go/errors"
+	"github.com/spiffe/spike-sdk-go/journal"
+)
+
+func TestRouteRecover_MemoryMode(t *testing.T) {
+	// Save original environment variables
+	originalStore := os.Getenv(env.NexusBackendStore)
+	defer func() {
+		if originalStore != "" {
+			_ = os.Setenv(env.NexusBackendStore, originalStore)
+		} else {
+			_ = os.Unsetenv(env.NexusBackendStore)
+		}
+	}()
+
+	// Set to memory mode
+	_ = os.Setenv(env.NexusBackendStore, "memory")
+
+	// Verify the environment is set correctly
+	if env.BackendStoreTypeVal() != env.Memory {
+		t.Fatal("Expected Memory backend store type")
+	}
+
+	// Create a test request
+	req := httptest.NewRequest(http.MethodPost, "/recover", nil)
+	w := httptest.NewRecorder()
+	audit := &journal.AuditEntry{}
+
+	// Call function
+	err := RouteRecover(w, req, audit)
+
+	// Recovery shards are useless for the in-memory backend; the route
+	// must reject the request explicitly rather than fail later with a
+	// misleading "not enough shards" internal error.
+	if err == nil {
+		t.Error("Expected an error in memory mode")
+		return
+	}
+	if !err.Is(sdkErrors.ErrDataInvalidInput) {
+		t.Errorf("Expected ErrDataInvalidInput, got: %v", err)
+	}
+	if w.Code != http.StatusBadRequest {
+		t.Errorf("Expected status %d, got: %d",
+			http.StatusBadRequest, w.Code)
+	}
+}

app/nexus/internal/route/operator/restore.go

@@ -63,9 +63,20 @@ func RouteRestore(
 	const fName = "routeRestore"
 	journal.AuditRequest(fName, r, audit, journal.AuditCreate)
 
+	// The in-memory backend keeps no persistent state, so there is
+	// nothing to restore. Reject the request explicitly: returning
+	// without a response body reads as "failed to communicate with
+	// SPIKE Nexus" on the Pilot side.
 	if env.BackendStoreTypeVal() == env.Memory {
-		log.Info(fName, "message", "skipping restoration: in-memory mode")
-		return nil
+		log.Warn(fName, "message", "rejecting restore: in-memory backend")
+		failErr := sdkErrors.ErrDataInvalidInput.Clone()
+		failErr.Msg = "restore is not applicable to the in-memory backend"
+		if respondErr := net.Fail(
+			reqres.RestoreResponse{}.BadRequest(), w, http.StatusBadRequest,
+		); respondErr != nil {
+			return failErr.Wrap(respondErr)
+		}
+		return failErr
 	}
 
 	request, err := net.ReadParseAndGuard[

app/nexus/internal/route/operator/restore_test.go

@@ -16,6 +16,7 @@ import (
 	"github.com/spiffe/spike-sdk-go/api/entity/v1/reqres"
 	"github.com/spiffe/spike-sdk-go/config/env"
 	"github.com/spiffe/spike-sdk-go/crypto"
+	sdkErrors "github.com/spiffe/spike-sdk-go/errors"
 
 	"github.com/spiffe/spike-sdk-go/journal"
 )
@@ -47,9 +48,18 @@ func TestRouteRestore_MemoryMode(t *testing.T) {
 	// Call function
 	err := RouteRestore(w, req, audit)
 
-	// Should return nil (no error) and skip processing in memory mode
-	if err != nil {
-		t.Errorf("Expected no error in memory mode, got: %v", err)
+	// Restore does not apply to the in-memory backend; the route must
+	// reject the request explicitly rather than return an empty 200.
+	if err == nil {
+		t.Error("Expected an error in memory mode")
+		return
+	}
+	if !err.Is(sdkErrors.ErrDataInvalidInput) {
+		t.Errorf("Expected ErrDataInvalidInput, got: %v", err)
+	}
+	if w.Code != http.StatusBadRequest {
+		t.Errorf("Expected status %d, got: %d",
+			http.StatusBadRequest, w.Code)
 	}
 }
 

app/spike/internal/cmd/operator/restore.go

@@ -5,8 +5,10 @@
 package operator
 
 import (
+	"bytes"
 	"context"
 	"encoding/hex"
+	"io"
 	"os"
 	"strconv"
 	"strings"
@@ -63,11 +65,8 @@ func newOperatorRestoreCommand(
 		Run: func(cmd *cobra.Command, args []string) {
 			spiffeid.IsPilotRestoreOrDie(SPIFFEID)
 
-			cmd.Println("(your input will be hidden as you paste/type it)")
-			cmd.Print("Enter recovery shard: ")
-			shard, readErr := term.ReadPassword(int(os.Stdin.Fd()))
+			shard, readErr := readShardInput(cmd)
 			if readErr != nil {
-				cmd.Println("") // newline after hidden input
 				cmd.PrintErrf("Error: %v\n", readErr)
 				return
 			}
@@ -167,3 +166,39 @@ func newOperatorRestoreCommand(
 
 	return restoreCmd
 }
+
+// readShardInput reads a recovery shard from standard input. When stdin
+// is a terminal, the input is hidden while typed. Otherwise, the shard
+// is read until EOF, which lets scripts (such as the bare-metal recovery
+// drill) drive `spike operator restore` non-interactively.
+//
+// In the non-interactive mode, the process supplying the shard holds a
+// copy of it too, so scripted restore should be reserved for development
+// environments and recovery drills.
+//
+// Parameters:
+//   - cmd: The Cobra command used for prompting.
+//
+// Returns:
+//   - []byte: The shard bytes with surrounding whitespace removed.
+//   - error: An error if reading standard input fails.
+func readShardInput(cmd *cobra.Command) ([]byte, error) {
+	fd := int(os.Stdin.Fd())
+
+	if term.IsTerminal(fd) {
+		cmd.Println("(your input will be hidden as you paste/type it)")
+		cmd.Print("Enter recovery shard: ")
+		shard, readErr := term.ReadPassword(fd)
+		cmd.Println("") // newline after hidden input
+		return shard, readErr
+	}
+
+	// A shard line is well under 4KB; the limit only guards against
+	// unbounded input.
+	data, readErr := io.ReadAll(io.LimitReader(os.Stdin, 4096))
+	if readErr != nil {
+		return nil, readErr
+	}
+
+	return bytes.TrimSpace(data), nil
+}

app/spike/internal/cmd/operator/restore_test.go

@@ -0,0 +1,79 @@
+//    \\ SPIKE: Secure your secrets with SPIFFE. — https://spike.ist/
+//  \\\\\ Copyright 2024-present SPIKE contributors.
+// \\\\\\\ SPDX-License-Identifier: Apache-2.0
+
+package operator
+
+import (
+	"os"
+	"strings"
+	"testing"
+
+	"github.com/spf13/cobra"
+)
+
+// TestReadShardInput_NonInteractive verifies that readShardInput reads a
+// shard from a non-terminal stdin (a pipe), which is what allows scripts
+// to drive `spike operator restore`.
+func TestReadShardInput_NonInteractive(t *testing.T) {
+	tests := []struct {
+		name  string
+		input string
+		want  string
+	}{
+		{
+			name:  "plain shard line",
+			input: "spike:1:" + strings.Repeat("ab", 32) + "\n",
+			want:  "spike:1:" + strings.Repeat("ab", 32),
+		},
+		{
+			name:  "surrounding whitespace is trimmed",
+			input: "  spike:2:" + strings.Repeat("cd", 32) + "  \n\n",
+			want:  "spike:2:" + strings.Repeat("cd", 32),
+		},
+		{
+			name:  "no trailing newline",
+			input: "spike:3:" + strings.Repeat("ef", 32),
+			want:  "spike:3:" + strings.Repeat("ef", 32),
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			r, w, pipeErr := os.Pipe()
+			if pipeErr != nil {
+				t.Fatalf("failed to create a pipe: %v", pipeErr)
+				return
+			}
+
+			original := os.Stdin
+			os.Stdin = r
+			t.Cleanup(func() {
+				os.Stdin = original
+				_ = r.Close()
+			})
+
+			if _, writeErr := w.WriteString(tt.input); writeErr != nil {
+				t.Fatalf("failed to write to the pipe: %v", writeErr)
+				return
+			}
+			if closeErr := w.Close(); closeErr != nil {
+				t.Fatalf("failed to close the pipe writer: %v", closeErr)
+				return
+			}
+
+			cmd := &cobra.Command{Use: "test"}
+
+			got, readErr := readShardInput(cmd)
+			if readErr != nil {
+				t.Fatalf("readShardInput() error = %v", readErr)
+				return
+			}
+
+			if string(got) != tt.want {
+				t.Errorf("readShardInput() = %q, want %q",
+					string(got), tt.want)
+			}
+		})
+	}
+}