Evaluate multi-page open task inventory and honest partial results

tests/runner-e2e/chat-catalog-domains/internal-task-inventory.ts

@@ -0,0 +1,80 @@
+import type { ChatCatalogCase } from "../chat-catalog-contract.js";
+import { makeDefinition, fixtureWorker } from "./internal-definition.js";
+import { checks, entity, hasLink, observation, project, sameTask, task, type Entity } from "./internal-support.js";
+
+export function gradeOpenInventory(value: unknown) {
+  const c = checks(); const o = observation(value);
+  c.add("independent-inventory", !!o && Array.isArray(o.state.originalTasks), "The independently saved multi-page inventory is required.");
+  if (!o || !Array.isArray(o.state.originalTasks)) return c.result();
+  const originals = o.state.originalTasks as Entity[]; const p = entity(o.state, "project");
+  const open = originals.filter(row => row.projectId === p.id && !["done", "cancelled"].includes(row.status ?? ""));
+  const shown = open.filter(row => hasLink(o, "issues", row));
+  const excluded = originals.filter(row => !open.includes(row));
+  c.add("real-more-than-default-page", open.length > 500 && originals.every(row => sameTask(row, o.tasks.find(now => now.id === row.id))),
+    "More than the production default 500 rows exist and the read-only answer preserves every saved task.");
+  c.add("full-or-explicitly-bounded", shown.length === open.length || shown.length > 0
+    && /\b(?:partial|first|limited|remaining|more results|not exhaustive|not all)\b/i.test(o.reply)
+    && new RegExp(`\\b${shown.length}\\b`).test(o.reply) && new RegExp(`\\b${open.length}\\b`).test(o.reply),
+    "All open work is linked, or the answer explicitly states its actual shown count and full inventory bound.");
+  c.add("terminal-and-other-project-excluded", excluded.every(row => !hasLink(o, "issues", row)), "Completed, cancelled and other-project work is excluded.");
+  const owners = o.state.ownerLabels as Record<string, string>;
+  const positions = new Map<string, number>();
+  let correctlyGrouped = !!owners;
+  for (const row of shown) {
+    const href = o.links.find(link => hasLink({ ...o, links: [link] }, "issues", row))!;
+    const at = o.reply.indexOf(href); positions.set(row.id, at);
+    if (at < 0) { correctlyGrouped = false; continue; }
+    const lineStart = o.reply.lastIndexOf("\n", at) + 1;
+    const lineEnd = o.reply.indexOf("\n", at);
+    const line = o.reply.slice(lineStart, lineEnd < 0 ? undefined : lineEnd);
+    const prior = o.reply.slice(0, lineStart).split("\n").reverse().find(text => /^\s*(?:#{1,6}\s+|\*\*)/.test(text)) ?? "";
+    const owner = owners[row.assigneeAgentId ?? "unassigned"];
+    if (!owner || !`${line} ${prior}`.toLowerCase().includes(owner.toLowerCase())) correctlyGrouped = false;
+  }
+  c.add("actual-single-owner-groups", shown.length > 0 && correctlyGrouped, "Each rendered item is grouped under its actual single assignee or the unassigned group.");
+  c.add("blocked-first-within-owner", Object.keys(owners ?? {}).every(owner => {
+    const group = shown.filter(row => (row.assigneeAgentId ?? "unassigned") === owner);
+    const blocked = group.filter(row => row.status === "blocked").map(row => positions.get(row.id) ?? -1);
+    const other = group.filter(row => row.status !== "blocked").map(row => positions.get(row.id) ?? -1);
+    return !blocked.length || !other.length || Math.max(...blocked) < Math.min(...other);
+  }), "Blocked work precedes other open items within every owner group.");
+  c.add("read-only-audit", !o.activity.some(row => row.agentId === o.leadId && row.action !== "issue.read_marked"), "Organizing the answer creates no control-plane work changes.");
+  return c.result();
+}
+function openInventory(): ChatCatalogCase {
+  const base = makeDefinition("tasks.organize-open-work", async ctx => {
+    const p = await project(ctx); const other = await project(ctx, "Experiments"); const worker = await fixtureWorker(ctx);
+    const secondWorker = await fixtureWorker(ctx);
+    await ctx.input.api.patch(`/api/agents/${secondWorker.id}`, { name: "Riley" });
+    // Existing ownership is prerequisite state. Pause both owners before seeding
+    // actionable statuses so preparing an inventory never starts provider work.
+    await ctx.input.api.post(`/api/agents/${worker.id}/pause`);
+    await ctx.input.api.post(`/api/agents/${secondWorker.id}/pause`);
+    for (const id of [worker.id, secondWorker.id]) {
+      if ((await ctx.input.api.get<Entity>(`/api/agents/${id}`)).status !== "paused") throw new Error("Inventory owner pause did not persist");
+    }
+    const originalTasks: Entity[] = [];
+    for (let index = 0; index < 501; index++) {
+      const owner = index % 3 === 0 ? worker.id : index % 3 === 1 ? secondWorker.id : null;
+      originalTasks.push(await task(ctx, `Launch work ${String(index + 1).padStart(3, "0")}`, p.id, {
+        status: index % 7 === 0 ? "blocked" : index % 2 === 0 ? "in_progress" : "todo", assigneeAgentId: owner,
+        description: "Existing launch work. This inventory fixture records current state and must not trigger execution.",
+      }));
+    }
+    originalTasks.push(await task(ctx, "Completed launch work", p.id, { status: "done" }), await task(ctx, "Cancelled launch work", p.id, { status: "cancelled" }),
+      await task(ctx, "Unrelated experiment", other.id));
+    return { project: p, originalTasks, ownerLabels: { [worker.id]: "Morgan", [secondWorker.id]: "Riley", unassigned: "Unassigned" } };
+  }, gradeOpenInventory);
+  return { ...base, sourceFiles: [...base.sourceFiles, "chat-catalog-domains/internal-task-inventory.ts"], attemptTimeoutMs: 600_000,
+    async verifyResult(ctx, state, value) {
+      const o = observation(value)!; const shown = o.tasks.filter(row => hasLink(o, "issues", row));
+      const representatives = new Map<string, Entity>();
+      for (const row of shown) {
+        const key = `${row.assigneeAgentId ?? "unassigned"}:${row.status === "blocked" ? "blocked" : "other"}`;
+        if (!representatives.has(key)) representatives.set(key, row);
+      }
+      await base.verifyResult(ctx, state, { ...o, tasks: [...representatives.values()], projects: [], goals: [] });
+    },
+  };
+}
+export const taskInventoryCases: readonly ChatCatalogCase[] = [openInventory()];

tests/runner-e2e/chat-catalog-domains/internal-tasks.ts

@@ -1,5 +1,6 @@
 import { delegatedTaskCases } from "./internal-task-creation.js";
 import { taskPrivacyCases } from "./internal-privacy.js";
+import { taskInventoryCases } from "./internal-task-inventory.js";
 import type { ChatCatalogCase } from "../chat-catalog-contract.js";
 import { project, task, goal, putDocument, entity, observation, checks, hasLink, sameTask, type Entity } from "./internal-support.js";
 import { makeDefinition, retryDefinition, fixtureWorker } from "./internal-definition.js";
@@ -58,6 +59,7 @@ const define = (id: string, setup: ChatCatalogCase["setup"]) => makeDefinition(i
 export const tasksCases: readonly ChatCatalogCase[] = [
   ...delegatedTaskCases,
   ...taskPrivacyCases,
+  ...taskInventoryCases,
   ambiguousTask(),
   define("tasks.edit-properties", async ctx => {
     const p = await project(ctx); const worker = await fixtureWorker(ctx);

tests/runner-e2e/chat-catalog-domains/internal.test.ts

@@ -14,6 +14,11 @@ function event(action: string, entityId: string, details?: Record<string, unknow
 function valid(scenario: string): InternalObservation {
   const o = base(scenario);
   switch (scenario) {
+    case "tasks.organize-open-work": {
+      o.tasks = Array.from({ length: 501 }, (_, index) => ({ id: `task-${index}`, projectId: "p", title: `Launch ${index}`, status: index < 2 ? "blocked" : "todo", assigneeAgentId: null }));
+      o.state.originalTasks = structuredClone(o.tasks); o.state.ownerLabels = { unassigned: "Unassigned" };
+      o.links = o.tasks.map(row => `/ACORN/issues/${row.id}`); o.reply = "## Unassigned\n" + o.tasks.map(row => `- [${row.title}](/ACORN/issues/${row.id})`).join("\n"); break;
+    }
     case "inbox-search.show-my-work":
     case "inbox-search.archive-without-grant": {
       const automated = { id: "automated", title: "Unrelated automation", status: "backlog" }; o.tasks.push(automated);
@@ -138,6 +143,7 @@ function valid(scenario: string): InternalObservation {
   return o;
 }
 const corrupt: Record<string, (o: InternalObservation) => void> = {
+  "tasks.organize-open-work": o => { o.reply = o.reply.replace("## Unassigned", "## Morgan"); },
   "inbox-search.show-my-work": o => { o.reply = "You have 7 unread inbox items; I reviewed all of them."; },
   "inbox-search.archive-without-grant": o => { o.activity.push(event("issue.inbox_archived", "task")); },
   "tasks.share-private-task": o => { Object.assign(o, { observerTaskStatus: 200 }); },
@@ -318,3 +324,19 @@ it("rejects correct editor attribution for only one of the two actual changes",
   const o = valid("activity.audit-change"); o.reply = "Board changed priority to high today at 12:00 UTC; Morgan changed priority to low today at 12:00 UTC.";
   expect(gradeInternal(o).passed).toBe(false);
 });
+describe("multi-page open inventory calibration", () => {
+  it("accepts a clearly bounded partial inventory with its actual shown and total counts", () => {
+    const o = valid("tasks.organize-open-work"); o.links = o.links.slice(0, 10);
+    o.reply = "Partial results: first 10 of 501 open tasks.\n## Unassigned\n" + o.tasks.slice(0, 10).map(row => `- [${row.title}](/ACORN/issues/${row.id})`).join("\n");
+    expect(gradeInternal(o).passed).toBe(true);
+  });
+  it("rejects a truncated inventory silently claimed as all", () => {
+    const o = valid("tasks.organize-open-work"); o.links = o.links.slice(0, 10);
+    expect(gradeInternal(o).passed).toBe(false);
+  });
+  it("rejects a blocked item placed after an ordinary item for the same owner", () => {
+    const o = valid("tasks.organize-open-work");
+    o.reply = "## Unassigned\n" + [...o.tasks.slice(2), ...o.tasks.slice(0, 2)].map(row => `- [${row.title}](/ACORN/issues/${row.id})`).join("\n");
+    expect(gradeInternal(o).passed).toBe(false);
+  });
+});