Review uncertain AI candidates and preserve unresolved business scope

apps/docs/scripts/auth-cli.ts

@@ -182,10 +182,10 @@ async function main() {
   assert.equal(researchResponse.status, 200, "authenticated research API");
   assert.ok(researchResponse.headers.get("cache-control")?.includes("no-store"));
   const research = await researchResponse.json();
-  assert.equal(research.companies.length, 1051);
+  assert.equal(research.companies.length, 1065);
   assert.equal(research.categories.length, 11);
-  assert.equal(research.evidence.length, 2697);
-  assert.equal(research.observations.length, 366);
+  assert.equal(research.evidence.length, 2764);
+  assert.equal(research.observations.length, 372);
   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/src/lib/ai-research.ts

@@ -48,7 +48,7 @@ export type ResearchCandidateBatch = {
   records: {
     id: string; name: string; url: string | null; categoryId: string;
     kind: "product" | "company" | "promotion"; profileStatus: "readable" | "unavailable" | "archive-only";
-    status: "pending-year" | "linked" | "existing-outside-cohort" | "outside-cohort" | "promotion";
+    status: "pending-year" | "pending-ai" | "linked" | "existing-outside-cohort" | "outside-cohort" | "promotion";
     companyId?: string; evidenceIds?: string[]; note: string;
   }[];
 };
@@ -125,13 +125,13 @@ export function validateResearch(snapshot: ResearchSnapshot) {
       if (!row.name.trim() || !row.note.trim() || !categories.has(row.categoryId) ||
           !["product", "company", "promotion"].includes(row.kind) ||
           !["readable", "unavailable", "archive-only"].includes(row.profileStatus) ||
-          !["pending-year", "linked", "existing-outside-cohort", "outside-cohort", "promotion"].includes(row.status) ||
+          !["pending-year", "pending-ai", "linked", "existing-outside-cohort", "outside-cohort", "promotion"].includes(row.status) ||
           (row.kind !== "promotion" && (!row.url || !evidenceUrl(row.url))) ||
           (row.url !== null && !evidenceUrl(row.url)) ||
           linked !== !!row.companyId || (row.companyId && !companies.has(row.companyId)) ||
           (row.kind === "promotion") !== (row.status === "promotion") ||
           (row.evidenceIds ?? []).some((id) => !evidence.has(id)) ||
-          (row.status === "outside-cohort" && !row.evidenceIds?.length))
+          (["outside-cohort", "pending-ai"].includes(row.status) && !row.evidenceIds?.length))
         throw new Error(`Invalid research candidate: ${row.id}`);
     }
   }

apps/docs/tests/research.test.ts

@@ -23,9 +23,9 @@ test("research revisions share the authentication database and preserve replay l
     for (let i = 0; i < 20; i++) consumeCode(config, "invalid");
     assert.equal(consumeCode(config, "invalid"), "limited");
     const initial = readResearchSnapshot();
-    assert.equal(initial.companies.length, 1051);
-    assert.equal(initial.evidence.length, 2697);
-    assert.equal(initial.observations.length, 366);
+    assert.equal(initial.companies.length, 1065);
+    assert.equal(initial.evidence.length, 2764);
+    assert.equal(initial.observations.length, 372);
     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");
@@ -64,8 +64,8 @@ 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, 170);
-  assert.equal(snapshot.reviews.filter((r) => r.scope === "dataset").length, 12);
+  assert.equal(snapshot.reviews.filter((r) => r.scope === "companies").length, 173);
+  assert.equal(snapshot.reviews.filter((r) => r.scope === "dataset").length, 13);
   assert.equal(snapshot.reviews.filter((r) => r.scope === "excluded-candidate").length, 44);
   const review = snapshot.reviews.find((r) => r.id === "R4")!;
   assert.deepEqual(reviewCompanyIds(snapshot, review), ["AI007", "AI013", "AI021", "AI034"]);
@@ -85,11 +85,11 @@ test("reviews link explicit companies without turning global or excluded cases i
 test("disclosure counts exclude client income, biographies, targets and duplicate milestones", () => {
   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), 52);
+  assert.equal(summary.reduce((n, c) => n + c.count, 0), 1065);
+  assert.equal(summary.reduce((n, c) => n + c.self, 0), 53);
   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), 978);
+  assert.equal(summary.reduce((n, c) => n + c.unknown, 0), 991);
   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"));
@@ -124,8 +124,8 @@ test("service run rates, legacy freight revenue and paid adoption do not become
 
 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, 83);
-  assert.equal(snapshot.companies.filter((c) => c.technicalReview?.level === "public-artifact").length, 18);
+  assert.equal(snapshot.companies.filter((c) => c.technicalReview).length, 85);
+  assert.equal(snapshot.companies.filter((c) => c.technicalReview?.level === "public-artifact").length, 19);
   assert.equal(snapshot.companies.filter((c) => c.technicalReview?.level === "independent-test").length, 1);
   assert.equal(snapshot.companies.find((c) => c.id === "AI051")?.revenueStatus, "unknown");
   assert.match(snapshot.companies.find((c) => c.id === "AI051")!.technicalReview!.testScope!, /Sep 2026/);
@@ -443,6 +443,55 @@ test("follow-up evidence can admit an uncertain directory candidate without rewr
   assert.throws(() => validateResearch(snapshot), /directory recheck/);
 });
 
+test("AI relevance follow-ups retain unresolved businesses and reuse existing company identities", () => {
+  const snapshot = structuredClone(seed) as ResearchSnapshot;
+  const batch = snapshot.candidateBatches!.find((b) => b.id === "YC-AI-UNCLEAR-1")!;
+  assert.equal(batch.complete, false);
+  assert.equal(batch.records.length, 24);
+  const original = coverageAudit.records.filter((r) => r.status === "ai-unclear").slice(0, 24);
+  assert.deepEqual(batch.records.map((r) => r.url), original.map((r) => r.source));
+  const linked = batch.records.filter((r) => r.status === "linked");
+  assert.equal(linked.length, 17);
+  assert.equal(linked.filter((r) => r.companyId!.startsWith("YC")).length, 14);
+  for (const [name, id] of [["Bud", "AI070"], ["Prism", "AI030"], ["Parametric", "AI014"]]) {
+    assert.equal(linked.find((r) => r.name === name)?.companyId, id);
+  }
+  assert.equal(snapshot.companies.filter((c) => c.profileOnly || c.directoryOrigin === "yc-2025-2026").length, 928);
+  for (const row of linked) {
+    const ycId = Number(row.id.replace("YC-AI1-", ""));
+    assert.equal(snapshot.directoryRechecks!.find((r) => r.ycId === ycId)?.companyId, row.companyId);
+  }
+  const pending = batch.records.filter((r) => r.status === "pending-ai");
+  assert.equal(pending.length, 7);
+  assert.ok(pending.every((r) => !r.companyId && r.evidenceIds!.length && /2025/.test(r.note)));
+  assert.match(pending.find((r) => r.name === "Better Auth")!.note, /AI.*模块/);
+  validateResearch(snapshot);
+  pending[0].companyId = "AI014";
+  assert.throws(() => validateResearch(snapshot), /research candidate/);
+  delete pending[0].companyId;
+  pending[0].evidenceIds = [];
+  assert.throws(() => validateResearch(snapshot), /research candidate/);
+});
+
+test("historical monthly revenue and customer examples do not enter recurring platform comparisons", () => {
+  const snapshot = seed as ResearchSnapshot;
+  const monthly = snapshot.observations.find((o) => o.id === "UNCLEAR1-O-SCALAR-MONTHLY")!;
+  assert.equal(monthly.value, 74000);
+  assert.equal(monthly.selection, "historical");
+  assert.equal(monthly.metric, undefined, "monthly revenue alone does not establish MRR or ARR");
+  assert.equal(monthly.comparisonGroup, null);
+  assert.equal(snapshot.companies.find((c) => c.id === "YC30514")!.revenueStatus, "self");
+  for (const id of ["AMBY-CLIENT", "NIXO-EXAMPLE", "AUTOSTEP-WASTE"]) {
+    assert.equal(snapshot.observations.find((o) => o.id === `UNCLEAR1-O-${id}`)!.selection, "excluded");
+  }
+  assert.ok(!platformIncomeSample(snapshot, { windowDays: null }).points.some((p) => p.observation.id.startsWith("UNCLEAR1-")));
+  const cua = snapshot.companies.find((c) => c.id === "YC30420")!;
+  assert.equal(cua.cat, "C03");
+  assert.equal(cua.revenueStatus, "unknown");
+  assert.match(cua.businessModel!, /自备计算机/);
+  assert.equal(cua.technicalReview!.level, "public-artifact");
+});
+
 test("Chinese discovery entries preserve unread articles and use separate primary evidence", () => {
   const snapshot = structuredClone(seed) as ResearchSnapshot;
   const batch = snapshot.candidateBatches!.find((b) => b.id === "JQ-2026-targeted-1")!;

content/research/ai-companies.sources.json

@@ -1,5 +1,5 @@
 {
-  "version": "2026-10-31-consolidated-1",
+  "version": "2026-10-31-ai-unclear-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。",
@@ -16,7 +16,7 @@
       "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。",
       "limitation": "批次不等于成立年；不代表全部公司",
-      "nextCheck": "已统一企业141、行业252、模型标签26、开发标签51、其余标签34项，共504项新增主营复核。当前928条原YC新增记录保留发现来源，其中基础项871条均已做主营人工复核（871条），57条深化个案。C05规则89保留27/移62，C05标签52保留10/移42；C07规则184保留135/移49，C07标签68保留39/移29；模型标签26保留17/移9，开发标签51保留16/移35，其余标签34保留16/移18。主营复核仍含C00/短简介/混合交付缺项，不等完整产品/技术/财务审计。Tasklet及Native新产品与旧业务链保留，发现审计不改；Freya、Epicenter、Stellon许可及公开材料/厂商基准范围分开。客户收入、估价、资产、节省、管线/意向、预订需求、旧公司收入与示例排除自身ARR；Locke服务run-rate不等ARR，Mixy历史MRR/捐赠、Isengard无期间收入分开。各批证据、观测、核查与主体/收费/人工成本缺项均保留；全部确认来源清单的候选范围、AI不明/成立冲突及公司财务技术缺项继续核，不能声称全来源研究完成。"
+      "nextCheck": "已统一企业141、行业252、模型标签26、开发标签51、其余标签34项，共504项新增主营复核。当前928条原YC新增记录保留发现来源，其中基础项871条均已做主营人工复核（871条），57条深化个案。C05规则89保留27/移62，C05标签52保留10/移42；C07规则184保留135/移49，C07标签68保留39/移29；模型标签26保留17/移9，开发标签51保留16/移35，其余标签34保留16/移18。主营复核仍含C00/短简介/混合交付缺项，不等完整产品/技术/财务审计。Tasklet及Native新产品与旧业务链保留，发现审计不改；Freya、Epicenter、Stellon许可及公开材料/厂商基准范围分开。客户收入、估价、资产、节省、管线/意向、预订需求、旧公司收入与示例排除自身ARR；Locke服务run-rate不等ARR，Mixy历史MRR/捐赠、Isengard无期间收入分开。各批证据、观测、核查与主体/收费/人工成本缺项均保留；全部确认来源清单的候选范围、AI不明/成立冲突及公司财务技术缺项继续核，不能声称全来源研究完成。 本轮另补核原AI不明池首24/192：14新增、3复用、7 AI主营待核（pending-ai），保留原审计不回写；本批外168项中OpenSEO已有较早复核，其余167项继续补核。Scalar Field历史月收入未声明MRR，客户收入/示例/成本排除，自备算力和软件中继收费分开。"
     },
     {
       "id": "S02",

content/research/ai-income-analysis.json

@@ -1,5 +1,5 @@
 {
-  "sourceVersion": "2026-10-31-consolidated-1",
+  "sourceVersion": "2026-10-31-ai-unclear-1",
   "asOf": "2026-10-31",
   "metric": "platform-mrr-usd",
   "sampleUnit": "product-or-business",

scripts/audit-ai-research.py

@@ -85,7 +85,7 @@ def candidate_register():
     all_ids = [r['id'] for b in batches for r in b['records']]
     assert len(set(all_ids)) == len(all_ids)
     lines = []
-    statuses = {'pending-year': '成立年待核', 'linked': '已关联公司',
+    statuses = {'pending-year': '成立年待核', 'pending-ai': 'AI主营待核', 'linked': '已关联公司',
                 'existing-outside-cohort': '已有关联，年份范围外',
                 'outside-cohort': '范围外或既有业务，不新增公司', 'promotion': '推广活动，非公司新增'}
     for batch in batches:
@@ -97,13 +97,13 @@ def candidate_register():
         assert sum(counts.values()) == len(rows)
         lines += ['', f"## {batch['sourceId']} 候选核对（{batch['id']}）", '',
             f"[发现页面]({batch['url']})，读取于{batch['accessedAt']}。{batch['scope']}", '',
-            f"本批{len(rows)}条：{counts['linked']}条关联公司、{counts['existing-outside-cohort']}条关联范围外对照、{counts['outside-cohort']}条范围外不新增公司、{counts['pending-year']}条成立年待核、{counts['promotion']}条推广活动。完整来源覆盖标记：{'已完成' if batch['complete'] else '未完成'}。主营或产品任务的初分类不计入主公司表数量，只有已关联的公司记录进入对应公司明细。候选与缺项随研究版本保存在文档站同一SQLite的研究元数据中。", '',
+            f"本批{len(rows)}条：{counts['linked']}条关联公司、{counts['existing-outside-cohort']}条关联范围外对照、{counts['outside-cohort']}条范围外不新增公司、{counts['pending-year']}条成立年待核、{counts['pending-ai']}条AI主营待核、{counts['promotion']}条推广活动。完整来源覆盖标记：{'已完成' if batch['complete'] else '未完成'}。主营或产品任务的初分类不计入主公司表数量，只有已关联的公司记录进入对应公司明细。候选与缺项随研究版本保存在文档站同一SQLite的研究元数据中。", '',
             '| 发现条目 | 主营或产品任务初分 | 核对状态 | 关联公司 | 核对证据 | 限制 |', '| --- | --- | --- | --- | --- | --- |']
         for row in rows:
             assert row['categoryId'] in categories
             assert row['status'] in statuses
             assert all(eid in evidence for eid in row.get('evidenceIds', []))
-            if row['status'] == 'outside-cohort':
+            if row['status'] in {'outside-cohort', 'pending-ai'}:
                 assert row.get('evidenceIds') and not row.get('companyId')
             name = f"[{row['name']}]({row['url']})" if row['url'] else row['name']
             company = '—'