fix(db): make filesystem lock intent explicit

src/shared/contracts/schemas-tasks.ts

@@ -32,6 +32,7 @@ export const taskInputSchema = z.object({
   description: z.string().default(""),
   assigneeAgentId: optionalId,
   workspaceId: optionalId,
+  requiresFilesystem: z.boolean().default(false),
   priority: z.enum(["critical", "high", "medium", "low"]).default("medium"),
   status: z.enum(["backlog", "todo", "in_progress", "in_review", "done", "blocked", "cancelled"]).default("todo"),
 });

src/shared/types/records-core.ts

@@ -157,12 +157,10 @@ export interface TaskRecord {
   taskType: string | null;
   requiresUserReview: boolean;
   /**
-   * Per-task filesystem lock policy (v52+). When `true` the
-   * orchestrator resolves a workspace and holds it exclusively for
-   * the duration of the run; when `false` the run executes with
-   * `lockMode='none'` and agents run in parallel on shared read-only
-   * context. Inferred from agent role at task creation and toggled
-   * explicitly by callers that know their task mutates the tree.
+   * Per-task filesystem lock policy (v52+). `true` makes the
+   * orchestrator resolve a workspace and hold it exclusively for
+   * the duration of the run. `false` lets a task with no explicit
+   * workspace binding execute with `lockMode='none'`.
    */
   requiresFilesystem: boolean;
   startedAt: string | null;
@@ -395,4 +393,3 @@ export interface ModelCapability {
   costPerMtokOut: number;
   updatedAt: string;
 }
-