Register all seven public Speedrun cohorts with verifiable boundaries

2026-11-01 01:54:41 +08:00
parent 17e9306c97
commit 51d6364ddd
6 changed files with 3624 additions and 12 deletions
@@ -184,7 +184,7 @@ async function main() {
  const research = await researchResponse.json();
  assert.equal(research.companies.length, 1253);
  assert.equal(research.categories.length, 11);
-  assert.equal(research.evidence.length, 4894);
+  assert.equal(research.evidence.length, 4909);
  assert.equal(research.observations.length, 491);
  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 });
@@ -124,7 +124,7 @@ 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, 1253);
-    assert.equal(initial.evidence.length, 4894);
+    assert.equal(initial.evidence.length, 4909);
    assert.equal(initial.observations.length, 491);
    const businessReviews = initial.reviews.filter((review) => review.type === businessModelReviewType);
    assert.equal(businessReviews.length, 7);
@@ -170,8 +170,8 @@ 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, 248);
-  assert.equal(snapshot.reviews.filter((r) => r.scope === "dataset").length, 147);
-  assert.equal(snapshot.reviews.filter((r) => r.scope === "excluded-candidate").length, 54);
+  assert.equal(snapshot.reviews.filter((r) => r.scope === "dataset").length, 149);
+  assert.equal(snapshot.reviews.filter((r) => r.scope === "excluded-candidate").length, 55);
  const review = snapshot.reviews.find((r) => r.id === "R4")!;
  assert.deepEqual(reviewCompanyIds(snapshot, review), ["AI007", "AI013", "AI021", "AI034"]);
  assert.deepEqual(reviewCompanyIds(snapshot, snapshot.reviews.find((r) => r.id === "CN-R-PSI")!), []);
@@ -630,6 +630,35 @@ test("the public SR007 roster preserves unique entries and Alven's older operato
    "a current category change must retain the unresolved historical product attribution");
 });

+test("public Speedrun cohort rosters preserve the finite boundary without inventing founding years", () => {
+  const snapshot = seed as ResearchSnapshot;
+  const rosters = snapshot.candidateBatches!.filter((batch) => /^SR00[1-7]-public-roster-2026-1$/.test(batch.id));
+  assert.equal(rosters.length, 7);
+  for (const [index, count] of [25, 25, 32, 42, 59, 59, 79].entries()) {
+    const batch = rosters.find((item) => item.id.startsWith(`SR${String(index + 1).padStart(3, "0")}-`))!;
+    assert.equal(batch.records.length, count);
+    assert.equal(batch.complete, false, "a complete directory page is not completed company-level research");
+  }
+  const rows = rosters.flatMap((batch) => batch.records);
+  assert.equal(rows.length, 321);
+  assert.equal(new Set(rows.map((row) => row.url)).size, 321);
+  const newlyRegistered = rosters.filter((batch) => !batch.id.startsWith("SR007-")).flatMap((batch) => batch.records);
+  assert.equal(newlyRegistered.length, 242);
+  assert.equal(newlyRegistered.filter((row) => row.status === "pending-year").length, 226);
+  assert.equal(newlyRegistered.filter((row) => row.status === "linked").length, 9);
+  assert.equal(newlyRegistered.filter((row) => row.status === "existing-outside-cohort").length, 1);
+  assert.equal(newlyRegistered.filter((row) => row.status === "outside-cohort").length, 6);
+  assert.ok(newlyRegistered.filter((row) => row.status === "outside-cohort").every((row) =>
+    row.profileStatus === "readable" && row.evidenceIds?.some((id) => id.startsWith("SRALL1-P-"))),
+  "old founding years require independent profile evidence, not cohort age");
+  const variants = newlyRegistered.filter((row) => row.name === "Variant");
+  assert.equal(variants.length, 2);
+  assert.equal(new Set(variants.map((row) => row.url)).size, 2, "same-name profiles must not be merged by name");
+  assert.equal(variants.filter((row) => row.companyId === "AI066").length, 1);
+  assert.ok(!snapshot.observations.some((row) => row.evidence.startsWith("SRALL1-")),
+    "roster discovery and outside-cohort claims do not create financial observations");
+});
+
 test("annual launch archives exclude older companies only with evidence and reuse existing records", () => {
  const snapshot = structuredClone(seed) as ResearchSnapshot;
  const batch = snapshot.candidateBatches!.find((b) => b.id === "PH-2025-visible-1")!;
@@ -657,7 +686,7 @@ test("unreconciled directory financials remain source reviews instead of income
  assert.ok(review.evidenceIds.includes("PH-TANA-DR"));
  assert.equal(reviewCompanyIds(snapshot, review).length, 0);
  assert.ok(!snapshot.observations.some((o) => o.evidence === "PH-TANA-DR"));
-  assert.equal(snapshot.candidateBatches!.flatMap((b) => b.records).filter((r) => r.status === "pending-year").length, 131);
+  assert.equal(snapshot.candidateBatches!.flatMap((b) => b.records).filter((r) => r.status === "pending-year").length, 357);
 });

 test("product pivots and founder histories do not become new company years or current revenue", () => {
@@ -1,5 +1,5 @@
 {
-  "version": "2026-11-01-sr007-followup-10",
+  "version": "2026-11-01-sr-public-rosters-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。",
@@ -3040,8 +3040,8 @@
        "speedrun.a16z.com"
      ],
      "initialAccess": "见具体证据访问记录",
-      "boundary": "原有定向个案保留。2026-11-01匿名普通HTML默认SR007载荷79条唯一ID/slug、75普通UUID链接，count79且next/previous空。当前71关联（17复用、54本系列新增）、Alven1年份/主体冲突、6AI主营待核、scaledrive2024超年1；本系列已读79独立档案（62补读＋17复用重读），全部位置有对应档案证据；复用原ID/观测，不新增重复公司。未执行站点脚本/私有接口；SR005/SR006参数仍返回SR007，不认其他批次或两年全量。Bric/Coherence/KIRA/Leonidas/Mecone/PhotonSpear当前AI主营未明。签约/cARR/管线/融资、客户资产/回款、旧事业业绩和界面财务不当已实现收入；live ARR、年度合同标签、年化服务和累计收入分别保留，财务日期/币种/净费不补造。软件订阅、额度充值、模型定制/运行、专业/人工托管、云存储、评测、硬件、金融手续费及旅行采购按交付拆开。所有公司交付/技术/收费缺项见关联证据和核查记录；公开说明、未发布功能、公开制品与独立测试分级。Sterling条款法人占位及Zoah的OPACITY残留只作补证，不能据此推造假。 本轮官网复核中Infragrid首页实验室数据/评测与条款Runbooks并列，当前暂C02但旧ARR不迁产品；Abliteration托管/API按C02，标题/正文口径不明仍保留。Codos/Proper明确含嵌入式工程，非纯自助软件假定。",
-      "nextCheck": "已完成17复用独立档案重读；继续核Alven2024主体/2025品牌沿革及6项AI主营。再建立其他批次可验证公开边界，不按请求参数认成功。公司层面补独立登记、真实合同/履约/取消、币种与财务日、经常性净费、模型/人工/资料授权/存储和渠道成本；重点包括Sterling年合同净费与法人、Vene个人/机构实付、Valent年化业务组成、Truli单次/订阅、Zoah训练退出/席位额度/运营者、Foundry公告/制品及Vound具体产品。其余公司核查缺项继续保留，不以本批可读证明全部真实。 Infragrid数据业务/旧Runbooks的产品沿革、合同与数据许可，Abliteration衍生模型权利/预收消费，Alia Prosieve及Proper GAIA/Virgo主体关系，Codos/Proper试点转续费及工程工时需补。",
+      "boundary": "2026-11-01公开前端匿名GET目录已取得SR001–SR007七批，分别25/25/32/42/59/59/79，共321个唯一公司ID/slug，声明/返回数一致且无下一页。只解析公开JSON，不执行JS、不登录、不调用投资者导出/联系或个人资料接口。原HTML默认SR007及其审计保留，公开GET的SR007 slug集合与其一致；此前普通页面?cohort参数无效不代表其他批次不可读。SR007的79独立档案已读：71关联、Alven1年份/主体冲突、6AI主营待核、scaledrive2024超年1。其余六批242位置仅10个已有档案URL精确复用（9两年公司、Rork2024对照），本轮另读Runware/Sloyd/Hedra/FLORA/Intangible/Agent Astra六项均旧年排除，226独立档案未读。当前321位置总80关联、1既有超年对照、7超年不新增、227成立/主体待核、6AI主营待核；不是所有公开档案均完成财务/注册/技术审计，也不含未公开/已移除/未来公司。原公司合同、业务转向、数据许可及技术缺项见逐条核查，不以完整目录充当完整公司研究。",
+      "nextCheck": "逐项读取新增六批226个未读候选档案，分清2025/2026成立、旧公司/品牌沿革及是否AI主营；10个复用档案保留既有证据并检查变化。同名Variant两链接先独立保留，不凭名字合并。继续核SR007的Alven主体/品牌和6项AI主营，以及已关联公司的登记、真实合同/履约/取消、币种/财务日、净经常性定义、人工/模型/硬件/数据成本。公开七批321为当前目录边界，不证明档案都已研究或两年无缺漏。",
      "limitation": "来源可读不等于声明已独立验证。",
      "rosterAudit": {
        "url": "https://speedrun.a16z.com/companies",
@@ -3055,7 +3055,121 @@
        "htmlUuidProfileLinks": 75,
        "htmlSha256": "32a8e1d9e6c9a2581077aed058767c46f03cdffdab9b0086ee9bd0c86094e152",
        "candidateBatchId": "SR007-public-roster-2026-1"
-      }
+      },
+      "publicRosterAudits": [
+        {
+          "cohort": "SR001",
+          "url": "https://speedrun.a16z.com/api/companies/companyparams?limit=100&offset=0&cohort=SR001",
+          "accessedAt": "2026-11-01",
+          "method": "公开前端实际使用的匿名GET，无Cookie/Authorization；只解析JSON不执行JS",
+          "httpStatus": 200,
+          "reportedCount": 25,
+          "recordCount": 25,
+          "uniquePublicIds": 25,
+          "uniqueSlugs": 25,
+          "next": null,
+          "previous": null,
+          "responseSha256": "9a4eb2c78fe918e31c88132f2052be26dfd07bddae6ebbe6ef5ea99f73897eea",
+          "candidateBatchId": "SR001-public-roster-2026-1",
+          "evidenceId": "SRALL1-SR001-ROSTER"
+        },
+        {
+          "cohort": "SR002",
+          "url": "https://speedrun.a16z.com/api/companies/companyparams?limit=100&offset=0&cohort=SR002",
+          "accessedAt": "2026-11-01",
+          "method": "公开前端实际使用的匿名GET，无Cookie/Authorization；只解析JSON不执行JS",
+          "httpStatus": 200,
+          "reportedCount": 25,
+          "recordCount": 25,
+          "uniquePublicIds": 25,
+          "uniqueSlugs": 25,
+          "next": null,
+          "previous": null,
+          "responseSha256": "2ac0887cb0d97ee6da4aa3f7b7d2ca4045b2640f0494e3be513c9ed6c05a79f8",
+          "candidateBatchId": "SR002-public-roster-2026-1",
+          "evidenceId": "SRALL1-SR002-ROSTER"
+        },
+        {
+          "cohort": "SR003",
+          "url": "https://speedrun.a16z.com/api/companies/companyparams?limit=100&offset=0&cohort=SR003",
+          "accessedAt": "2026-11-01",
+          "method": "公开前端实际使用的匿名GET，无Cookie/Authorization；只解析JSON不执行JS",
+          "httpStatus": 200,
+          "reportedCount": 32,
+          "recordCount": 32,
+          "uniquePublicIds": 32,
+          "uniqueSlugs": 32,
+          "next": null,
+          "previous": null,
+          "responseSha256": "8b205f1b6e58aec25b5bf0d39ae5bb29d6d0b10cc7144606792ea134cb7a345f",
+          "candidateBatchId": "SR003-public-roster-2026-1",
+          "evidenceId": "SRALL1-SR003-ROSTER"
+        },
+        {
+          "cohort": "SR004",
+          "url": "https://speedrun.a16z.com/api/companies/companyparams?limit=100&offset=0&cohort=SR004",
+          "accessedAt": "2026-11-01",
+          "method": "公开前端实际使用的匿名GET，无Cookie/Authorization；只解析JSON不执行JS",
+          "httpStatus": 200,
+          "reportedCount": 42,
+          "recordCount": 42,
+          "uniquePublicIds": 42,
+          "uniqueSlugs": 42,
+          "next": null,
+          "previous": null,
+          "responseSha256": "eae528acacf9a52bb001b4a2e42c774662849dcb130dfc009488cc86ec8b7475",
+          "candidateBatchId": "SR004-public-roster-2026-1",
+          "evidenceId": "SRALL1-SR004-ROSTER"
+        },
+        {
+          "cohort": "SR005",
+          "url": "https://speedrun.a16z.com/api/companies/companyparams?limit=100&offset=0&cohort=SR005",
+          "accessedAt": "2026-11-01",
+          "method": "公开前端实际使用的匿名GET，无Cookie/Authorization；只解析JSON不执行JS",
+          "httpStatus": 200,
+          "reportedCount": 59,
+          "recordCount": 59,
+          "uniquePublicIds": 59,
+          "uniqueSlugs": 59,
+          "next": null,
+          "previous": null,
+          "responseSha256": "82a94a083f45f8c61d72667f196b8b7afb7a03d25c386cfdb6cc212a21579ac8",
+          "candidateBatchId": "SR005-public-roster-2026-1",
+          "evidenceId": "SRALL1-SR005-ROSTER"
+        },
+        {
+          "cohort": "SR006",
+          "url": "https://speedrun.a16z.com/api/companies/companyparams?limit=100&offset=0&cohort=SR006",
+          "accessedAt": "2026-11-01",
+          "method": "公开前端实际使用的匿名GET，无Cookie/Authorization；只解析JSON不执行JS",
+          "httpStatus": 200,
+          "reportedCount": 59,
+          "recordCount": 59,
+          "uniquePublicIds": 59,
+          "uniqueSlugs": 59,
+          "next": null,
+          "previous": null,
+          "responseSha256": "baa82a040e3a8466241a86fdc3c681ffd114ba55f9a16c3f73448227468726c7",
+          "candidateBatchId": "SR006-public-roster-2026-1",
+          "evidenceId": "SRALL1-SR006-ROSTER"
+        },
+        {
+          "cohort": "SR007",
+          "url": "https://speedrun.a16z.com/api/companies/companyparams?limit=100&offset=0&cohort=SR007",
+          "accessedAt": "2026-11-01",
+          "method": "公开前端实际使用的匿名GET，无Cookie/Authorization；只解析JSON不执行JS",
+          "httpStatus": 200,
+          "reportedCount": 79,
+          "recordCount": 79,
+          "uniquePublicIds": 79,
+          "uniqueSlugs": 79,
+          "next": null,
+          "previous": null,
+          "responseSha256": "9a9d437fbd16ed477f2844b84ac1ad037ad3e0c9215013c7a9a90126569ab915",
+          "candidateBatchId": "SR007-public-roster-2026-1",
+          "evidenceId": "SRALL1-SR007-ROSTER"
+        }
+      ]
    },
    {
      "id": "S15",
@@ -1,5 +1,5 @@
 {
-  "sourceVersion": "2026-11-01-sr007-followup-10",
+  "sourceVersion": "2026-11-01-sr-public-rosters-1",
  "asOf": "2026-11-01",
  "metric": "platform-mrr-usd",
  "sampleUnit": "product-or-business",
@@ -4,7 +4,7 @@ from collections import Counter
 import json
 from pathlib import Path
 import sys
-from urllib.parse import urlsplit
+from urllib.parse import parse_qs, urlsplit

 ROOT = Path(__file__).resolve().parents[1]
 DATA = json.loads((ROOT / 'content/research/ai-companies.seed.json').read_text())
@@ -49,6 +49,27 @@ def source_register():
            assert roster['defaultCohort'] == 'SR007' and roster['next'] is None
            assert len(batch['records']) == roster['recordCount'] == roster['reportedCount']
            assert len({row['url'] for row in batch['records']}) == roster['recordCount']
+        if source['id'] == 'S14' and source.get('publicRosterAudits'):
+            audits = source['publicRosterAudits']
+            assert {a['cohort'] for a in audits} == {f'SR{i:03d}' for i in range(1, 8)}
+            assert len(audits) == 7
+            profile_urls = []
+            for audit in audits:
+                batch = next(b for b in DATA['candidateBatches'] if b['id'] == audit['candidateBatchId'])
+                query = parse_qs(urlsplit(audit['url']).query)
+                assert urlsplit(audit['url']).netloc == 'speedrun.a16z.com'
+                assert query == {'limit': ['100'], 'offset': ['0'], 'cohort': [audit['cohort']]}
+                assert urlsplit(audit['url']).path == '/api/companies/companyparams'
+                assert audit['httpStatus'] == 200 and audit['accessedAt'] <= DATA['asOf']
+                assert audit['next'] is None and audit['previous'] is None
+                assert len(batch['records']) == audit['recordCount'] == audit['reportedCount']
+                assert audit['uniquePublicIds'] == audit['uniqueSlugs'] == audit['recordCount']
+                assert len(audit['responseSha256']) == 64 and all(c in '0123456789abcdef' for c in audit['responseSha256'])
+                assert batch['sourceId'] == 'S14' and not batch['complete']
+                proof = next(e for e in DATA['evidence'] if e['id'] == audit['evidenceId'])
+                assert proof['url'] == audit['url'] and proof['companyId'] is None
+                profile_urls.extend(row['url'] for row in batch['records'])
+            assert len(profile_urls) == len(set(profile_urls)), 'duplicate profiles across public cohorts'
        if source['url']:
            assert urlsplit(source['url']).scheme in {'http', 'https'} and urlsplit(source['url']).netloc
        for host in source['hosts']: