npm - job-pro - Versions diffs - 0.7.0 → 0.7.2 - Mend

job-pro 0.7.0 → 0.7.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (24) hide show

package/dist/adapter.js +17 -0
package/dist/alibaba.js +2 -2
package/dist/baichuan.js +38 -123
package/dist/bilibili.js +43 -0
package/dist/byd.js +333 -112
package/dist/cambricon.js +379 -39
package/dist/deepseek.js +373 -25
package/dist/galaxyuniversal.js +375 -23
package/dist/horizonrobotics.js +40 -61
package/dist/iflytek.js +299 -57
package/dist/index.js +55 -50
package/dist/megvii.js +387 -150
package/dist/mihoyo.js +241 -67
package/dist/moonshot.js +387 -54
package/dist/oppo.js +212 -56
package/dist/pdd.js +278 -129
package/dist/sensetime.js +47 -183
package/dist/sf.js +249 -36
package/dist/stepfun.js +333 -105
package/dist/vivo.js +291 -41
package/dist/wecruit.js +385 -0
package/dist/weibo.js +259 -88
package/dist/zerooneai.js +40 -37
package/package.json +3 -2

package/dist/pdd.js CHANGED Viewed

@@ -1,7 +1,7 @@
 // Thin client for 拼多多 (PDD / Pinduoduo) campus-recruiting portal at careers.pinduoduo.com.
 //
 // ============================================================
-// API discovery (probed 2026-05, Next.js bundle 4SRcl1zmo3aLPqQ_wcJ6T):
+// API discovery (re-probed 2026-05; bundles under pfile.pddpic.com/ei-pub):
 //
 // The frontend is a Next.js SPA hosted at careers.pinduoduo.com/campus/.
 // All XHR calls go through a sourceSDK.wrappedRequest helper (module 96211)
@@ -9,52 +9,56 @@
 //
 //   https://careers.pinduoduo.com/api/careers/<relative-url>
 //
-// ---- Probed endpoints and their accessibility (2026-05) ----
+// ---- PUBLIC endpoints (no auth required, verified 2026-05) ----
 //
-// PUBLIC (no auth required):
-//   POST /api/careers/api/campus/moment/list
-//        Body: { pageSize:<int>, page:<int> }
-//        Response: { success:true, result:{ list:[MomentItem], total:<int> } }
-//        MomentItem: { guid, momentTitle, momentLabel:[string], publishDate:<ms>, topFlag:bool }
+// POST /api/careers/api/recruit/position/list
+//      Body: { pageSize:<int>, page:<int> }   (keyword/recruitType IGNORED server-side)
+//      Response: { success:true, result:{ list:[Position], total:"<int>" } }
+//      Position: { id (uuid), name, code, workLocation, workLocationName,
+//                  job (eng key), jobName (zh), releaseTime (ms),
+//                  jobDuty (full JD HTML/text), labelList[], recruitTypeName }
+//      Pagination: page size is FIXED at 10 regardless of pageSize param.
+//      Total returned across all pages (e.g. 30 for grad, 7 for intern).
 //
-//   POST /api/careers/api/campus/moment/detail
-//        Body: { guid:<string> }
-//        Response: { success:true, result:{ guid, momentTitle, momentContent:<html> } }
+// POST /api/careers/api/recruit/position/train/list
+//      Same shape as /list but only intern (实习生 / 2027届研发实习生) positions.
 //
-//   POST /api/careers/api/recruit/qa/common/list
-//        Body: {} (no params needed)
-//        Response: { success:true, result:[{ question, questionCode }] }
-//        Note: only 2 FAQ items are currently surfaced publicly (2026-05).
+// POST /api/careers/api/recruit/position/detail/type
+//      Body: {} → returns the job-type dictionary
+//      Result: [{job:"technology", jobName:"技术"}, {job:"general", jobName:"职能"},
+//               {job:"product", jobName:"产品"}, {job:"language", jobName:"语言"},
+//               {job:"market", jobName:"市场营销"}, {job:"visual", jobName:"视觉类"},
+//               {job:"investment", jobName:"运营"}, {job:"vegetable", jobName:"区域业务"}]
 //
-//   POST /api/careers/api/campus/careers/enum
-//        Body: {} → { success:true, result:{ enumMap:{} } }  (empty in public response)
+// POST /api/careers/api/campus/moment/list      (notices / 公告)
+//      Body: { pageSize:<int>, page:<int> }
+//      Response: { success:true, result:{ list:[MomentItem], total:<int> } }
+//      MomentItem: { guid, momentTitle, momentLabel:[string], publishDate:<ms>, topFlag:bool }
 //
-// AUTH-REQUIRED (401 Unauthorized without valid login token):
-//   POST /api/careers/api/recruit/position/queryPosition   ← main job list
-//   POST /api/careers/api/recruit/position/querySecondPosition
-//   POST /api/careers/api/recruit/site/query
-//   POST /api/careers/api/campus/area/full/list
-//   POST /api/careers/api/campus/education/major/query
-//   POST /api/careers/api/recruit/qa/list
-//   POST /api/careers/api/recruit/queryByShortLink
+// POST /api/careers/api/campus/moment/detail
+//      Body: { guid:<string> }
+//      Response: { success:true, result:{ guid, momentTitle, momentContent:<html> } }
 //
-// ANTI-BOT BLOCKED (403 without Anti-Content token):
-//   The old /api/* paths (without /api/careers/ prefix) return 403
-//   from PDD's yak/openresty CDN layer. These require a dynamically
-//   generated Anti-Content header produced by PDD's JS risk-control SDK.
+// POST /api/careers/api/campus/trip/list        (校招行程 / on-campus events)
+//      Body: {} → { tripList:null|[...], explainTrip:{ recruitmentTripType, tripContent:<html> } }
 //
-// ---- Conclusion ----
-// The position search/detail APIs require a registered PDD account login
-// token. There is no publicly accessible job listing endpoint. This adapter
-// provides honest stubs for positions and implements real data for notices
-// (moment/list) and the FAQ (qa/common/list).
+// POST /api/careers/api/recruit/qa/common/list  (FAQ)
+//      Body: {} → [{ question, questionCode }] (only ~2 items publicly)
+//
+// POST /api/careers/api/campus/careers/enum     (enum map, empty for anon)
+//
+// ---- AUTH-REQUIRED endpoints (HTTP 401 without login token) ----
+//   /api/recruit/position/queryPosition          (rich search with all filters)
+//   /api/recruit/position/querySecondPosition
+//   /api/recruit/site/query, /campus/area/full/list, /campus/education/major/query
+//   /api/recruit/qa/list, /api/recruit/queryByShortLink
 //
 // ---- RecruitType taxonomy (from JS bundle, probed 2026-05) ----
 //   headquarters    = 管培生 (headquarters management trainee)
 //   region          = 区域业务管培生 (regional business management trainee)
 //   technical_session = 技术专场 (technical session / R&D)
 //   warehouse_trainee = 仓储类管培生 (warehouse management trainee)
-//   yunhu_plan      = 云弧计划 (LLM / elite tech talent program, equivalent to ByteDance 顶尖)
+//   yunhu_plan      = 云弧计划 (LLM / elite tech talent program)
 //   intern          = 实习生 (intern, 2027届)
 //
 // ---- Campus batches active at time of probing (2026-05) ----
@@ -63,6 +67,17 @@
 //   云弧计划 (elite LLM/AI talent program, 2026) → route /grad?recruitType=yunhu_plan
 //
 // ============================================================
+// IMPLEMENTATION NOTES
+//   - searchPositions: uses /api/recruit/position/list (grad track). Server
+//     ignores `keyword` and `pageSize`; we filter client-side by keyword and
+//     slice the response to honour the requested pageSize. Server returns
+//     fixed pages of 10.
+//   - fetchAllPositions: paginates through both /list and /train/list to
+//     surface every public position.
+//   - fetchPositionDetail: there is no per-position detail endpoint; the
+//     /list response already includes the full jobDuty. We scan the list to
+//     locate the matching record by id.
+// ============================================================
 import { extractResumeSignals, scoreOverlap, checkResume } from "./tencent.js";
 export { checkResume };
 const API_ROOT = "https://careers.pinduoduo.com/api/careers";
@@ -120,70 +135,158 @@ async function call(path, body) {
         message: ok ? "ok" : (payload.errorMsg || `errorCode ${payload.errorCode}`),
     };
 }
+const DETAIL_URL = (id) => `${GRAD_PAGE}?id=${encodeURIComponent(id)}`;
+function summarizePosition(raw) {
+    const id = raw.id ?? "";
+    return {
+        post_id: id,
+        title: raw.name ?? "",
+        project: raw.jobName ?? "",
+        recruit_label: raw.recruitTypeName ?? "",
+        bgs: (raw.labelList ?? []).join(" / "),
+        work_cities: raw.workLocationName ?? raw.workLocation ?? "",
+        apply_url: id ? DETAIL_URL(id) : GRAD_PAGE,
+    };
+}
+// Server returns 10 items/page regardless of pageSize. Treat that as the upstream chunk.
+const UPSTREAM_PAGE_SIZE = 10;
+function asNumber(value) {
+    if (typeof value === "number")
+        return value;
+    if (typeof value === "string") {
+        const n = parseInt(value, 10);
+        return Number.isFinite(n) ? n : 0;
+    }
+    return 0;
+}
 // ---------- searchPositions ----------
-// PDD's position search API (queryPosition) requires a logged-in user token.
-// All calls return HTTP 401 without a valid session. This stub documents the
-// attempted endpoint and returns ok:false so callers can degrade gracefully.
+// Uses the public /api/recruit/position/list endpoint. The server ignores
+// `keyword` and `pageSize` (always returns 10 items/page in releaseTime order),
+// so we paginate upstream, then filter+slice client-side to honour the
+// caller-requested keyword/pageSize.
 export async function searchPositions(opts = {}) {
     const pageSize = Math.max(1, Math.min(100, opts.pageSize ?? 20));
     const page = Math.max(1, opts.page ?? 1);
-    const keyword = (opts.keyword ?? "").trim().slice(0, 60);
-    const payload = {
-        pageSize,
-        page,
-        ...(keyword ? { keyword } : {}),
-        ...(opts.recruitType ? { recruitType: opts.recruitType } : {}),
-    };
-    // Attempt the real API — expected to return 401 without auth token.
-    const response = await call("/api/recruit/position/queryPosition", payload);
-    // If the server starts accepting unauthenticated requests in the future,
-    // this block will handle the data. For now it always falls through to the
-    // stub error return below.
-    if (response.ok && response.data) {
-        // Future-proofing: if API becomes public, parse and return positions.
-        return {
-            ok: true,
-            source: "careers.pinduoduo.com",
-            query: payload,
-            page,
-            page_size: pageSize,
-            total: 0,
-            positions: [],
-            note: "Position data returned by server but field mapping not yet implemented — please file an issue.",
-        };
+    const keyword = (opts.keyword ?? "").trim().slice(0, 60).toLowerCase();
+    const includeIntern = opts.recruitType === "intern";
+    // Walk upstream until we have enough filtered rows to satisfy the requested page.
+    const collected = [];
+    let upstreamTotal = 0;
+    const need = page * pageSize;
+    const maxUpstreamPages = 10; // safety cap (10 pages × 10 = 100 positions)
+    const endpoints = includeIntern
+        ? ["/api/recruit/position/train/list"]
+        : ["/api/recruit/position/list"];
+    for (const path of endpoints) {
+        for (let p = 1; p <= maxUpstreamPages; p++) {
+            const response = await call(path, {
+                pageSize: UPSTREAM_PAGE_SIZE,
+                page: p,
+            });
+            if (!response.ok || !response.data) {
+                // Surface auth-style failures as ok:false.
+                if (collected.length === 0) {
+                    return {
+                        ok: false,
+                        source: "careers.pinduoduo.com",
+                        query: { pageSize, page, keyword: keyword || undefined, recruitType: opts.recruitType },
+                        page,
+                        page_size: pageSize,
+                        total: 0,
+                        positions: [],
+                        message: response.message,
+                        apply_at: GRAD_PAGE,
+                    };
+                }
+                break;
+            }
+            upstreamTotal = Math.max(upstreamTotal, asNumber(response.data.total));
+            const batch = response.data.list ?? [];
+            if (!batch.length)
+                break;
+            for (const item of batch) {
+                if (!keyword) {
+                    collected.push(item);
+                }
+                else {
+                    const hay = `${item.name ?? ""} ${item.recruitTypeName ?? ""} ${item.jobName ?? ""} ${item.workLocationName ?? ""} ${item.jobDuty ?? ""}`.toLowerCase();
+                    if (hay.includes(keyword))
+                        collected.push(item);
+                }
+            }
+            // Continue paginating until we have enough or exhausted the upstream pool.
+            if (collected.length >= need && p * UPSTREAM_PAGE_SIZE >= upstreamTotal)
+                break;
+            if (batch.length < UPSTREAM_PAGE_SIZE)
+                break;
+        }
     }
-    const isAuth = response.httpStatus === 401;
+    const start = (page - 1) * pageSize;
+    const slice = collected.slice(start, start + pageSize);
     return {
-        ok: false,
+        ok: true,
         source: "careers.pinduoduo.com",
-        query: payload,
+        query: { pageSize, page, keyword: keyword || undefined, recruitType: opts.recruitType },
         page,
         page_size: pageSize,
-        total: 0,
-        positions: [],
-        message: isAuth
-            ? "PDD position API requires a registered account login token (HTTP 401). " +
-                "The endpoint POST /api/careers/api/recruit/position/queryPosition was reached " +
-                "but rejected unauthenticated requests. Apply at: " + GRAD_PAGE
-            : response.message,
+        total: keyword ? collected.length : upstreamTotal || collected.length,
+        positions: slice.map(summarizePosition),
         apply_at: GRAD_PAGE,
     };
 }
 // ---------- fetchAllPositions ----------
+// Walks both grad (/list) and intern (/train/list) tracks until exhaustion.
 export async function fetchAllPositions(opts = {}) {
-    // Delegates to searchPositions; inherits the auth stub.
-    const result = await searchPositions(opts);
+    const pageSize = Math.max(1, Math.min(100, opts.pageSize ?? 50));
+    const maxPages = Math.max(1, opts.maxPages ?? 20);
+    const keyword = (opts.keyword ?? "").trim().slice(0, 60).toLowerCase();
+    const bucket = [];
+    const seen = new Set();
+    let total = 0;
+    for (const path of ["/api/recruit/position/list", "/api/recruit/position/train/list"]) {
+        let pathTotal = 0;
+        for (let p = 1; p <= maxPages; p++) {
+            const response = await call(path, {
+                pageSize: UPSTREAM_PAGE_SIZE,
+                page: p,
+            });
+            if (!response.ok || !response.data)
+                break;
+            pathTotal = Math.max(pathTotal, asNumber(response.data.total));
+            const batch = response.data.list ?? [];
+            if (!batch.length)
+                break;
+            for (const item of batch) {
+                const id = item.id ?? "";
+                if (!id || seen.has(id))
+                    continue;
+                seen.add(id);
+                if (keyword) {
+                    const hay = `${item.name ?? ""} ${item.jobName ?? ""} ${item.recruitTypeName ?? ""} ${item.workLocationName ?? ""} ${item.jobDuty ?? ""}`.toLowerCase();
+                    if (!hay.includes(keyword))
+                        continue;
+                }
+                bucket.push(item);
+            }
+            if (batch.length < UPSTREAM_PAGE_SIZE)
+                break;
+        }
+        total += pathTotal;
+        if (bucket.length >= pageSize * maxPages)
+            break;
+    }
     return {
-        ok: result.ok,
+        ok: true,
         source: "careers.pinduoduo.com",
-        total: result.total ?? 0,
-        fetched: 0,
-        positions: [],
-        message: result.message,
+        total: keyword ? bucket.length : total || bucket.length,
+        fetched: bucket.length,
+        positions: bucket.map(summarizePosition),
         apply_at: GRAD_PAGE,
     };
 }
 // ---------- fetchPositionDetail ----------
+// There is no per-position detail endpoint; the list response already carries
+// the full jobDuty. We scan grad + intern lists for the matching uuid.
 export async function fetchPositionDetail(postId) {
     const id = (postId ?? "").trim();
     if (!id) {
@@ -193,28 +296,55 @@ export async function fetchPositionDetail(postId) {
             message: "post_id is required",
         };
     }
-    // There is no public detail endpoint; queryPosition (which lists with detail)
-    // also requires auth. Return a stub with the closest apply URL.
+    for (const path of ["/api/recruit/position/list", "/api/recruit/position/train/list"]) {
+        for (let p = 1; p <= 10; p++) {
+            const response = await call(path, {
+                pageSize: UPSTREAM_PAGE_SIZE,
+                page: p,
+            });
+            if (!response.ok || !response.data)
+                break;
+            const batch = response.data.list ?? [];
+            if (!batch.length)
+                break;
+            const hit = batch.find((row) => row.id === id);
+            if (hit) {
+                const summary = summarizePosition(hit);
+                return {
+                    ok: true,
+                    source: "careers.pinduoduo.com",
+                    ...summary,
+                    description: hit.jobDuty ?? "",
+                    requirements: "",
+                    work_cities: hit.workLocationName ? [hit.workLocationName] : [],
+                    release_time: hit.releaseTime
+                        ? new Date(hit.releaseTime).toISOString().slice(0, 10)
+                        : "",
+                    code: hit.code ?? "",
+                    labels: hit.labelList ?? [],
+                };
+            }
+            if (batch.length < UPSTREAM_PAGE_SIZE)
+                break;
+        }
+    }
     return {
         ok: false,
         source: "careers.pinduoduo.com",
         post_id: id,
-        description: "",
-        requirements: "",
-        work_cities: [],
         apply_url: GRAD_PAGE,
-        message: "PDD position detail API requires account auth (HTTP 401). " +
-            `Visit ${GRAD_PAGE} to browse positions.`,
+        message: "Position not found in current public list. The position may have been " +
+            "closed or moved to an auth-only track. " +
+            `Visit ${GRAD_PAGE} to browse current openings.`,
     };
 }
-// ---------- fetchDictionaries ----------
-// The /api/careers/api/campus/careers/enum endpoint is publicly accessible
-// but returns an empty enumMap. Recruit types are documented in the header
-// comment above (extracted from the JS bundle).
 export async function fetchDictionaries() {
-    const response = await call("/api/campus/careers/enum", {});
+    const [enumResp, typeResp] = await Promise.all([
+        call("/api/campus/careers/enum", {}),
+        call("/api/recruit/position/detail/type", {}),
+    ]);
     return {
-        ok: response.ok,
+        ok: enumResp.ok || typeResp.ok,
         source: "careers.pinduoduo.com",
         api_host: API_ROOT,
         verified_at: new Date().toISOString(),
@@ -226,12 +356,14 @@ export async function fetchDictionaries() {
             { value: "technical_session", label: "技术专场", note: "Technical session / R&D" },
             { value: "warehouse_trainee", label: "仓储类管培生", note: "Warehouse management trainee" },
             { value: "yunhu_plan", label: "云弧计划", note: "LLM/AI elite talent program (≈ ByteDance 顶尖实习)" },
+            { value: "intern", label: "实习生", note: "Intern (2027届)" },
         ],
+        job_types: (typeResp.data ?? []).map((t) => ({ value: t.job ?? "", label: t.jobName ?? "" })),
         current_batch: "2026届春季校招 / 2027届研发实习生",
         grad_page: GRAD_PAGE,
         intern_page: INTERN_PAGE,
-        enum_map: response.data?.enumMap ?? {},
-        note: "Position search requires account login; dict shows static bundle data + live enum endpoint.",
+        enum_map: enumResp.data?.enumMap ?? {},
+        note: "Position list is public via /api/recruit/position/list — see header comment.",
     };
 }
 function formatNotice(item) {
@@ -375,11 +507,11 @@ export async function findNoticesByQuestion(question, opts = {}) {
     };
 }
 // ---------- matchResume ----------
-// Mirrors bytedance.ts/tencent.ts algo but acknowledges that position data
-// is unavailable. Returns extracted signals and the apply URL for the user
-// to manually browse matching positions.
+// Pulls every public position (grad + intern), scores each against the resume's
+// extracted terms and city preferences, and returns the top-N matches.
 export async function matchResume(text, opts = {}) {
     const { terms, cities } = extractResumeSignals(text ?? "");
+    const topN = Math.max(1, opts.topN ?? 5);
     if (!terms.length) {
         return {
             ok: false,
@@ -388,46 +520,63 @@ export async function matchResume(text, opts = {}) {
             preview: (text ?? "").slice(0, 120),
         };
     }
-    // We cannot search positions (auth required), so we return a best-effort
-    // recommendation: which PDD recruit type best matches the resume signals,
-    // with a direct link to the appropriate page.
-    const isTechnical = terms.some((t) => ["python", "java", "go", "golang", "c++", "cpp", "rust", "typescript",
-        "javascript", "react", "vue", "spring", "pytorch", "tensorflow",
-        "kubernetes", "docker", "linux", "llm", "rag", "transformer",
-        "后端", "前端", "算法", "推荐", "搜索", "大模型", "测试", "运维", "安全"].includes(t.toLowerCase()));
-    const isLLM = terms.some((t) => ["llm", "rag", "transformer", "bert", "gpt", "diffusion", "大模型", "强化学习", "多模态"].includes(t.toLowerCase()));
-    const suggested = [];
-    if (isLLM) {
-        suggested.push({
-            recruitType: "yunhu_plan",
-            label: "云弧计划 (LLM Elite)",
-            url: `${GRAD_PAGE}?recruitType=yunhu_plan`,
-            reason: "LLM/AI signals detected: " + terms.filter((t) => ["llm", "rag", "transformer", "bert", "gpt", "diffusion", "大模型"].includes(t.toLowerCase())).slice(0, 3).join(", "),
-        });
+    const all = await fetchAllPositions({ maxPages: 10, pageSize: UPSTREAM_PAGE_SIZE });
+    if (!all.ok) {
+        return {
+            ok: false,
+            source: "careers.pinduoduo.com",
+            extracted_terms: terms,
+            city_preferences: cities,
+            matches: [],
+            message: all.message ?? "failed to fetch positions",
+            apply_at: GRAD_PAGE,
+        };
     }
-    if (isTechnical) {
-        suggested.push({
-            recruitType: "technical_session",
-            label: "技术专场 (R&D)",
-            url: `${GRAD_PAGE}?recruitType=technical_session`,
-            reason: "Technical signals detected: " + terms.slice(0, 3).join(", "),
-        });
+    // Score against the raw positions we already have via fetchAllPositions's caller.
+    // Re-pull as raw to retain jobDuty for scoring.
+    const rawCorpus = [];
+    for (const path of ["/api/recruit/position/list", "/api/recruit/position/train/list"]) {
+        for (let p = 1; p <= 10; p++) {
+            const response = await call(path, {
+                pageSize: UPSTREAM_PAGE_SIZE,
+                page: p,
+            });
+            if (!response.ok || !response.data)
+                break;
+            const batch = response.data.list ?? [];
+            if (!batch.length)
+                break;
+            rawCorpus.push(...batch);
+            if (batch.length < UPSTREAM_PAGE_SIZE)
+                break;
+        }
     }
-    suggested.push({
-        recruitType: "headquarters",
-        label: "管培生 (Management Trainee)",
-        url: `${GRAD_PAGE}?recruitType=headquarters`,
-        reason: "General campus track",
-    });
+    const scored = [];
+    for (const raw of rawCorpus) {
+        const hay = `${raw.name ?? ""} ${raw.jobName ?? ""} ${raw.recruitTypeName ?? ""} ${raw.jobDuty ?? ""}`.toLowerCase();
+        const matched = terms.filter((t) => hay.includes(t.toLowerCase()));
+        const overlap = scoreOverlap(hay, terms, cities).score;
+        const city_match = cities.length === 0 ? false :
+            cities.some((c) => (raw.workLocationName ?? raw.workLocation ?? "").includes(c));
+        if (!matched.length && !city_match)
+            continue;
+        const score = overlap * 100 + matched.length * 5 + (city_match ? 8 : 0);
+        scored.push({ raw, score, matched_terms: matched, city_match });
+    }
+    scored.sort((a, b) => b.score - a.score);
+    const matches = scored.slice(0, topN).map((s) => ({
+        ...summarizePosition(s.raw),
+        score: s.score,
+        matched_terms: s.matched_terms,
+        city_match: s.city_match,
+    }));
     return {
         ok: true,
         source: "careers.pinduoduo.com",
         extracted_terms: terms,
         city_preferences: cities,
-        note: "PDD position search API requires account auth — cannot rank individual JDs. " +
-            "Suggested tracks are derived from resume signals against static recruit-type taxonomy.",
-        suggested_tracks: suggested.slice(0, opts.topN ?? 3),
-        matches: [],
+        total_scanned: rawCorpus.length,
+        matches,
         apply_at: GRAD_PAGE,
     };
 }