research: review industry workflows and historical ARR boundaries

apps/docs/scripts/auth-cli.ts

@@ -184,8 +184,8 @@ async function main() {
   const research = await researchResponse.json();
   assert.equal(research.companies.length, 1051);
   assert.equal(research.categories.length, 11);
-  assert.equal(research.evidence.length, 1858);
-  assert.equal(research.observations.length, 237);
+  assert.equal(research.evidence.length, 1896);
+  assert.equal(research.observations.length, 242);
   assert.equal((await request("/api/research/ai-companies", { method: "POST", headers })).status, 405);
   const researchPage = await request("/docs/other-research/ai-company-research/overview", { headers });
   assert.equal(researchPage.status, 200);

apps/docs/tests/research.test.ts

@@ -24,8 +24,8 @@ test("research revisions share the authentication database and preserve replay l
     assert.equal(consumeCode(config, "invalid"), "limited");
     const initial = readResearchSnapshot();
     assert.equal(initial.companies.length, 1051);
-    assert.equal(initial.evidence.length, 1858);
-    assert.equal(initial.observations.length, 237);
+    assert.equal(initial.evidence.length, 1896);
+    assert.equal(initial.observations.length, 242);
     assert.deepEqual(initial.candidateBatches, seed.candidateBatches, "candidate gaps persist in the same database metadata");
     assert.deepEqual(initial.directoryRechecks, seed.directoryRechecks, "follow-up classifications persist without rewriting the discovery audit");
     assert.deepEqual(readResearchSnapshot(), initial, "reopening does not duplicate records");
@@ -64,7 +64,7 @@ test("research validation rejects broken evidence and unsafe source links", () =
 
 test("reviews link explicit companies without turning global or excluded cases into companies", () => {
   const snapshot = structuredClone(seed) as ResearchSnapshot;
-  assert.equal(snapshot.reviews.filter((r) => r.scope === "companies").length, 115);
+  assert.equal(snapshot.reviews.filter((r) => r.scope === "companies").length, 118);
   assert.equal(snapshot.reviews.filter((r) => r.scope === "dataset").length, 12);
   assert.equal(snapshot.reviews.filter((r) => r.scope === "excluded-candidate").length, 44);
   const review = snapshot.reviews.find((r) => r.id === "R4")!;
@@ -86,10 +86,10 @@ test("disclosure counts exclude client income, biographies, targets and duplicat
   const snapshot = seed as ResearchSnapshot;
   const summary = categorySummary(snapshot);
   assert.equal(summary.reduce((n, c) => n + c.count, 0), 1051);
-  assert.equal(summary.reduce((n, c) => n + c.self, 0), 43);
+  assert.equal(summary.reduce((n, c) => n + c.self, 0), 45);
   assert.equal(summary.reduce((n, c) => n + c.estimate, 0), 2);
   assert.equal(summary.reduce((n, c) => n + c.platform, 0), 19);
-  assert.equal(summary.reduce((n, c) => n + c.unknown, 0), 987);
+  assert.equal(summary.reduce((n, c) => n + c.unknown, 0), 985);
   assert.equal(snapshot.companies.find((c) => c.name === "Rork")?.revenueStatus, "unknown");
   assert.equal(snapshot.companies.find((c) => c.name === "Result")?.revenueStatus, "unknown");
   assert.ok(snapshot.observations.filter((o) => o.selection === "signal").every((o) => o.scope === "公司整体"));
@@ -101,7 +101,7 @@ test("disclosure counts exclude client income, biographies, targets and duplicat
 
 test("technical reviews keep accessible materials separate from tested capabilities", () => {
   const snapshot = structuredClone(seed) as ResearchSnapshot;
-  assert.equal(snapshot.companies.filter((c) => c.technicalReview).length, 59);
+  assert.equal(snapshot.companies.filter((c) => c.technicalReview).length, 61);
   assert.equal(snapshot.companies.filter((c) => c.technicalReview?.level === "public-artifact").length, 9);
   assert.equal(snapshot.companies.filter((c) => c.technicalReview?.level === "independent-test").length, 1);
   assert.equal(snapshot.companies.find((c) => c.id === "AI051")?.revenueStatus, "unknown");

content/research/ai-companies.sources.json

@@ -1,5 +1,5 @@
 {
-  "version": "2026-10-31-1",
+  "version": "2026-10-31-industry-1",
   "scopeConfirmedAt": "2026-10-30",
   "acceptance": "以确认的公开来源清单为边界，逐条核对并标明缺项。",
   "provenance": "S01—S12恢复自已提交用户的《AI公司研究_来源与分类模板》来源清单；S13—S16单列后续已使用的补充证据渠道，不将使用次数当作全量覆盖。",
@@ -15,7 +15,7 @@
       ],
       "initialAccess": "官网目录可访问",
       "boundary": "已读取八个2025—2026批次镜像候选池1428条，再核YC官方页面；不声称覆盖全部YC历年或未公开公司。",
-      "nextCheck": "原新增目录中31条深化为个案并保留来源；当前基础项393条主营分类人工复核。本轮23项C06规则初分已读，22项保留、Bluma转创意制作，8项官网资料与Native政府接口缺项分开。软件/人工托管/制作代理及广告预算拆分；Native官网旧经营与YC2026成立冲突，当前year=null，原发现审计保留；2M ARR仅交叉核对。继续核C05/C07及其他非人工资料、主体、技术和财务，不把主营复核当完整来源覆盖。",
+      "nextCheck": "原新增目录33条深化个案并保留来源；当前基础项416条主营人工复核。本分支新增原C07首批25项复核：16保留、9按现买家与交付迁移，Flott/Adentris历史ARR单列。Linewise现训练数据，实体承包、专业服务和个人应用分开。Wayline附载Second旧帖未充分核主体延续，不强改2025年或继承前业务。原C07尚余159项规则初分；本快照原C05仍有89项规则初分，继续补充人工结果；继续其他非人工资料与来源缺项，不把分类当财务/技术全量验证。",
       "limitation": "批次不等于成立年；不代表全部公司"
     },
     {

content/research/ai-income-analysis.json

@@ -1,5 +1,5 @@
 {
-  "sourceVersion": "2026-10-31-sales-models-1",
+  "sourceVersion": "2026-10-31-industry-1",
   "asOf": "2026-10-31",
   "metric": "platform-mrr-usd",
   "sampleUnit": "product-or-business",