feat(auth): integrate authentication with installation lifecycle

backend/src/auth/auth.ts

@@ -82,12 +82,6 @@ export function authenticateSession(deps: AuthDependencies): preHandlerHookHandl
     const snapshot = captureAuthConfigSnapshot(request, deps.authentication);
     if (isPublicRoute(request)) return;
 
-    const operator = loopbackMaintenancePrincipal(request);
-    if (operator) {
-      request.principal = operator;
-      return;
-    }
-
     if (legacy) {
       await legacy(request, reply);
       if (reply.sent || !STATE_CHANGING_METHODS.has(request.method)) return;
@@ -144,15 +138,6 @@ export function authenticateSession(deps: AuthDependencies): preHandlerHookHandl
   };
 }
 
-function loopbackMaintenancePrincipal(request: FastifyRequest): PrincipalContext | undefined {
-  if (request.ip !== "127.0.0.1" && request.ip !== "::1" && request.ip !== "::ffff:127.0.0.1") return undefined;
-  if (singleHeader(request.headers["x-thoth-principal-issuer"]) !== "tht"
-    || singleHeader(request.headers["x-thoth-principal-subject"]) !== "tht-maintenance"
-    || singleHeader(request.headers["x-thoth-principal-display-name"]) !== "Tht maintenance"
-    || singleHeader(request.headers["x-thoth-is-admin"]) !== "1") return undefined;
-  return upstreamPrincipal(request.headers);
-}
-
 export function requireCsrf(request: FastifyRequest, reply: FastifyReply): true | FastifyReply {
   const expectedOrigin = request.authPublicOrigin;
   const token = request.authSessionToken;

backend/src/operator-command.ts

@@ -0,0 +1,123 @@
+/**
+ * Installation-scoped host operations that must not impersonate an HTTP administrator.
+ * This command runs only through `docker compose exec core`; it never accepts credentials,
+ * headers, paths, or arbitrary code from the caller.
+ */
+import { pathToFileURL } from "node:url";
+import { join } from "node:path";
+import { loadConfig, type AppConfig } from "./config.js";
+import { rolesToPermissions } from "./auth/config.js";
+import type { PrincipalContext } from "./auth/principal.js";
+import { createPiModelLister } from "./pi/list-models.js";
+import { createPiManagement } from "./pi/management.js";
+import { effectiveSettings } from "./routes/settings.js";
+import { MaintenanceBarrier } from "./runtime/maintenance-gate.js";
+import { loadSettings } from "./settings/settings-store.js";
+import { ThtRunner, type SessionRow } from "./tht/tht-runner.js";
+import { WorkspaceRegistry } from "./workspaces/registry.js";
+import { WorkspaceSecretStore } from "./workspaces/secret-store.js";
+
+type OperatorAction = "maintenance-activate" | "maintenance-deactivate" | "maintenance-status"
+  | "session-inventory" | "workflow-doctor" | "pi-options" | "pi-test" | "effective-settings";
+
+const lifecyclePrincipal: PrincipalContext = {
+  issuer: "tht-operator-command",
+  subject: "installation-lifecycle",
+  displayName: "Installation lifecycle",
+  roles: ["admin"],
+  permissions: rolesToPermissions(["admin"]),
+  isAdmin: true,
+};
+
+function operatorRunner(config: AppConfig): ThtRunner {
+  const workspaceSecretStore = new WorkspaceSecretStore({
+    root: config.workspaceSecretStoreRoot,
+    runtimeRoot: config.workspaceSecretRuntimeRoot,
+    installationId: config.workspaceRegistry.installationId,
+  });
+  return new ThtRunner({
+    thtBin: config.thtBin,
+    harnessDir: config.harnessDir,
+    configPath: process.env.THT_CONFIG ?? "config/tht.yaml",
+    dataRoot: config.dataRoot,
+    runtimeSnapshotRoot: join(config.workspaceRegistry.root, "snapshots", "runtime"),
+    secretRoots: config.workspaceRegistry.secretRoots,
+    secretsFile: config.secretsFile,
+    secretFiles: config.secretFiles,
+    workspaceSecretStore,
+    semanticRuntime: {
+      internalQdrantUrl: config.internalQdrantUrl,
+      internalEmbeddingUrl: config.internalEmbeddingUrl,
+      internalEmbeddingModel: config.internalEmbeddingModel,
+      internalEmbeddingDimensions: config.internalEmbeddingDimensions,
+    },
+  }).withPrincipal(lifecyclePrincipal);
+}
+
+async function sessionInventory(config: AppConfig): Promise<Array<Pick<SessionRow, "status" | "archived">>> {
+  const registry = new WorkspaceRegistry(config.workspaceRegistry);
+  const revisions = await registry.listRetainedSnapshots();
+  const runner = operatorRunner(config);
+  const sessions = new Map<string, SessionRow>();
+  for (const revision of revisions) {
+    for (const session of await runner.sessionList(revision.snapshotPath)) sessions.set(session.id, session);
+  }
+  return [...sessions.values()].map(({ status, archived }) => ({ status, archived: archived === true }));
+}
+
+async function workflowDiagnostics(config: AppConfig): Promise<{ ready: true; workspaces: number }> {
+  const registry = new WorkspaceRegistry(config.workspaceRegistry);
+  const revisions = await registry.listRetainedSnapshots();
+  if (revisions.length === 0) throw new Error("workflow diagnostics unavailable");
+  const runner = operatorRunner(config);
+  for (const revision of revisions) {
+    const result = await runner.run(["doctor", "--json"], revision.snapshotPath);
+    let payload: unknown;
+    try {
+      payload = JSON.parse(result.stdout);
+    } catch {
+      throw new Error("workflow diagnostics failed");
+    }
+    if (
+      result.code !== 0 || !payload || typeof payload !== "object"
+      || (payload as { ok?: unknown }).ok !== true
+    ) throw new Error("workflow diagnostics failed");
+  }
+  return { ready: true, workspaces: revisions.length };
+}
+
+export async function runOperatorAction(
+  action: OperatorAction,
+  config: AppConfig,
+): Promise<unknown> {
+  if (action.startsWith("maintenance-")) {
+    const barrier = new MaintenanceBarrier(config.maintenanceFile);
+    if (action === "maintenance-activate") await barrier.activate();
+    if (action === "maintenance-deactivate") barrier.deactivate();
+    return barrier.status();
+  }
+  if (action === "session-inventory") return await sessionInventory(config);
+  if (action === "workflow-doctor") return await workflowDiagnostics(config);
+  if (action === "effective-settings") return effectiveSettings(config, loadSettings(config));
+  const service = createPiManagement(config, { listModels: createPiModelLister(config) });
+  if (action === "pi-options") return await service.options();
+  if (action === "pi-test") return await service.test();
+  throw new Error("unsupported operator action");
+}
+
+async function main(): Promise<void> {
+  const action = process.argv[2] as OperatorAction | undefined;
+  if (!action || ![
+    "maintenance-activate", "maintenance-deactivate", "maintenance-status", "session-inventory",
+    "workflow-doctor", "pi-options", "pi-test", "effective-settings",
+  ].includes(action)) throw new Error("invalid operator action");
+  const result = await runOperatorAction(action, loadConfig(process.env));
+  process.stdout.write(`${JSON.stringify(result)}\n`);
+}
+
+if (process.argv[1] && import.meta.url === pathToFileURL(process.argv[1]).href) {
+  void main().catch(() => {
+    process.stderr.write("operator command failed\n");
+    process.exitCode = 2;
+  });
+}

backend/src/server.ts

@@ -1,10 +1,10 @@
 import { buildApp, type AppWithAuthSessionStore } from "./app.js";
 import { loadConfig } from "./config.js";
 import { formatStartupFailure } from "./startup-error.js";
-const config = loadConfig(process.env);
-const app = buildApp(config) as AppWithAuthSessionStore;
 
 async function start(): Promise<void> {
+  const config = loadConfig(process.env);
+  const app = buildApp(config) as AppWithAuthSessionStore;
   const sessions = app.thothiiAuthSessionStore;
   if (sessions) {
     await sessions.prune();

backend/src/startup-error.ts

@@ -4,9 +4,15 @@ const STARTUP_CAUSES = new Set([
   "workspace_registry_invalid",
 ]);
 
+const EXACT_CAUSES = new Map([
+  ["authentication configuration is invalid", "auth_config_invalid"],
+  ["authentication session state is invalid", "auth_session_store_invalid"],
+  ["workspace registry configuration is invalid", "workspace_registry_invalid"],
+]);
+
 /** Return one bounded machine cause; never include the original error text or stack. */
 export function formatStartupFailure(error: unknown): string {
   const message = error instanceof Error ? error.message : "";
-  const cause = STARTUP_CAUSES.has(message) ? message : "startup_unknown";
+  const cause = STARTUP_CAUSES.has(message) ? message : EXACT_CAUSES.get(message) ?? "startup_unknown";
   return `backend startup failed: ${cause}`;
 }