[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_154":155,"_156":157},"routes/_app.x.$handle.$name",{"_95":94,"_32":45,"_103":85,"_27":104,"_13":14,"_105":-5,"_11":106,"_107":108,"_109":110,"_111":112,"_113":114,"_20":-5,"_115":-5,"_116":-5,"_117":25,"_118":119,"_120":121,"_122":-5,"_123":-5,"_124":125},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":-5,"_18":-5},"project",{"_43":44,"_32":45,"_38":46,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":53,"_54":-5,"_18":55,"_56":-5,"_57":25,"_58":-5,"_59":-5,"_60":61,"_62":63,"_64":-5,"_65":-5,"_66":-5,"_67":-5,"_68":69,"_70":69,"_71":72,"_73":72,"_74":72,"_75":69,"_76":69,"_77":78,"_79":80,"_81":82,"_83":25},"version",3,"changes",[19],"aiChatInfo","userId",{"_20":21,"_22":23,"_24":25,"_26":25,"_27":28,"_29":28},"file",{"_30":31,"_32":28,"_33":37,"_35":36,"_38":39,"_40":14,"_41":42,"_24":25},"previous",{"_30":31,"_32":28,"_33":34,"_35":36},"deleted",false,"deletedHere","path","main.ts","latestPath","fileId","b04a19be-3f50-11f1-82b1-42b51c65c3df","name","fileType","script","content","// ============================================================================\n// SEC 8-K Event Scraper — v1\n// Captures Item-code disclosures + keyword hits from recent 8-K filings.\n// Philosophy matches 10-K scraper v4: capture everything, don't filter,\n// let the downstream LLM decide what matters.\n// ============================================================================\n\nconst HEADERS = {\n \"User-Agent\": \"AlexKrasiuk alex@myrunbook.com\",\n \"Accept-Encoding\": \"gzip, deflate\",\n};\n\nconst CONTEXT_CHARS = 800;\nconst MAX_FILINGS = 20; // how many recent 8-Ks to scan\nconst LOOKBACK_DAYS = 365; // only look at filings from last year\nconst MAX_HITS_RETURNED = 150;\n\n// Item codes that map to B2B sales signals get a score boost.\n// We KEEP all items regardless — boost is informational only.\nconst HIGH_SIGNAL_ITEMS: Record = {\n \"1.01\": { title: \"Entry into Material Definitive Agreement\", boost: 5 },\n \"1.02\": { title: \"Termination of Material Definitive Agreement\", boost: 4 },\n \"2.01\": {\n title: \"Completion of Acquisition or Disposition of Assets\",\n boost: 6,\n },\n \"2.02\": { title: \"Results of Operations and Financial Condition\", boost: 3 },\n \"2.05\": {\n title: \"Costs Associated with Exit or Disposal Activities\",\n boost: 10,\n },\n \"2.06\": { title: \"Material Impairments\", boost: 8 },\n \"3.02\": { title: \"Unregistered Sales of Equity Securities\", boost: 2 },\n \"5.02\": { title: \"Departure/Appointment of Officers or Directors\", boost: 7 },\n \"5.03\": { title: \"Amendments to Articles/Bylaws\", boost: 2 },\n \"7.01\": { title: \"Regulation FD Disclosure\", boost: 2 },\n \"8.01\": { title: \"Other Events\", boost: 2 },\n};\n\n// Same keyword list as 10-K scraper v4 for consistency\nconst KEYWORDS = [\n // AI / ML\n \"artificial intelligence\",\n \"AI agent\",\n \"AI agents\",\n \"generative AI\",\n \"gen AI\",\n \"GenAI\",\n \"machine learning\",\n \"large language model\",\n \"LLM\",\n \"AI-enabled\",\n \"AI-powered\",\n \"AI initiative\",\n \"AI technology\",\n \"AI tool\",\n\n // Headcount / workforce\n \"headcount reduction\",\n \"headcount reductions\",\n \"workforce reduction\",\n \"reduction in force\",\n \"reduction-in-force\",\n \"workforce optimization\",\n \"restructuring\",\n \"severance\",\n \"severance charges\",\n \"layoff\",\n \"layoffs\",\n \"job cut\",\n \"overhead reduction\",\n \"workforce transformation\",\n \"reorganization\",\n\n // Back office / ops\n \"back office\",\n \"back-office\",\n \"shared services\",\n \"administrative function\",\n \"administrative functions\",\n \"centralize\",\n \"centralization\",\n \"offshore\",\n \"outsourc\",\n\n // Automation\n \"process automation\",\n \"robotic process automation\",\n \"RPA\",\n \"automate\",\n \"automated\",\n \"automation\",\n \"intelligent automation\",\n \"workflow automation\",\n\n // Transformation\n \"digital transformation\",\n \"technology transformation\",\n \"digital initiative\",\n \"modernize\",\n \"modernization\",\n\n // Efficiency\n \"process optimization\",\n \"operational efficiency\",\n \"cost savings\",\n \"cost reduction\",\n \"margin expansion\",\n \"margin improvement\",\n \"efficiency initiative\",\n \"efficiency program\",\n\n // Manual work\n \"manual processes\",\n \"streamline\",\n \"streamlined\",\n\n // Field ops\n \"dispatch\",\n \"scheduling\",\n \"route optimization\",\n \"field service\",\n \"work order\",\n \"call center\",\n \"contact center\",\n\n // Financial pressure\n \"margin pressure\",\n \"labor cost\",\n \"labor shortage\",\n];\n\n// ============================================================================\n// Ticker cache (24h) — shared pattern from v4\n// ============================================================================\n\nlet tickerCache: { data: any; ts: number } | null = null;\nconst CACHE_TTL_MS = 24 * 60 * 60 * 1000;\n\nasync function getTickers(): Promise {\n if (tickerCache && Date.now() - tickerCache.ts < CACHE_TTL_MS) {\n return tickerCache.data;\n }\n const data = JSON.parse(\n await secFetch(\"https://www.sec.gov/files/company_tickers.json\"),\n );\n tickerCache = { data, ts: Date.now() };\n return data;\n}\n\n// ============================================================================\n// SEC fetch with 429 retry\n// ============================================================================\n\nasync function secFetch(url: string, retries = 3): Promise {\n for (let i = 0; i < retries; i++) {\n const resp = await fetch(url, { headers: HEADERS });\n if (resp.ok) return resp.text();\n if (resp.status === 429) {\n await new Promise((r) => setTimeout(r, 1000 * (i + 1)));\n continue;\n }\n throw new Error(`SEC fetch failed: ${resp.status} for ${url}`);\n }\n throw new Error(`SEC fetch failed after ${retries} retries for ${url}`);\n}\n\n// ============================================================================\n// HTML stripping — same as v4\n// ============================================================================\n\nfunction stripHtml(html: string): string {\n let text = html.replace(/]*>.*?<\\/script>/gis, \" \");\n text = text.replace(/]*>.*?<\\/style>/gis, \" \");\n text = text.replace(/<\\/?ix:[^>]+>/gi, \" \");\n text = text.replace(/<[^>]+>/g, \" \");\n text = text\n .replace(/&/g, \"&\").replace(/</g, \"<\").replace(/>/g, \">\")\n .replace(/"/g, '\"').replace(/'/g, \"'\").replace(/ /g, \" \")\n .replace(/&#\\d+;/g, \" \").replace(/&[a-zA-Z]+;/g, \" \");\n text = text.replace(/\\s+/g, \" \").trim();\n return text;\n}\n\nfunction normalize(s: string): string {\n return s.toLowerCase().replace(/[^a-z0-9\\s]/g, \"\").replace(/\\s+/g, \" \")\n .trim();\n}\n\n// ============================================================================\n// 8-K Item parsing\n// 8-Ks are structured by \"Item X.XX\" markers. We find all markers,\n// then each item's content is the text between its marker and the next.\n// ============================================================================\n\ntype ItemBlock = {\n item_code: string;\n item_title: string;\n content: string;\n start: number; // char offset in text — used to map keyword hits to items\n};\n\nfunction parseItems(text: string): ItemBlock[] {\n // Find all \"Item X.XX\" markers\n const markers: { code: string; idx: number }[] = [];\n const markerRe = /Item\\s+(\\d{1,2}\\.\\d{2})\\b/g;\n let m;\n while ((m = markerRe.exec(text)) !== null) {\n markers.push({ code: m[1], idx: m.index });\n }\n\n // De-dup consecutive identical codes (appears in TOC + body)\n const deduped: { code: string; idx: number }[] = [];\n for (const mk of markers) {\n const prev = deduped[deduped.length - 1];\n // If we see the same item again within 200 chars, it's a TOC artifact\n if (prev && prev.code === mk.code && mk.idx - prev.idx < 200) continue;\n deduped.push(mk);\n }\n\n const items: ItemBlock[] = [];\n for (let i = 0; i < deduped.length; i++) {\n const start = deduped[i].idx;\n const end = i + 1 < deduped.length ? deduped[i + 1].idx : text.length;\n const block = text.slice(start, end).trim();\n\n // Try to extract the title (everything between the item number and\n // the first sentence-ending period or a meaningful content start)\n const titleMatch = block.match(\n /^Item\\s+\\d+\\.\\d+\\.?\\s+([A-Z][^\\n]{5,200}?)(?=\\s{2,}|\\.\\s+[A-Z]|\\bOn\\s|\\bThe\\s|\\bPursuant)/,\n );\n const title = titleMatch\n ? titleMatch[1].trim().replace(/\\s+/g, \" \")\n : HIGH_SIGNAL_ITEMS[deduped[i].code]?.title || \"Unknown\";\n\n items.push({\n item_code: deduped[i].code,\n item_title: title,\n content: block,\n start,\n });\n }\n return items;\n}\n\nfunction findContainingItem(\n items: ItemBlock[],\n matchIdx: number,\n): ItemBlock | null {\n // Find the item whose range contains matchIdx\n for (let i = 0; i < items.length; i++) {\n const start = items[i].start;\n const end = i + 1 < items.length\n ? items[i + 1].start\n : Number.MAX_SAFE_INTEGER;\n if (matchIdx >= start && matchIdx < end) return items[i];\n }\n return null;\n}\n\n// ============================================================================\n// Scoring — informational only\n// ============================================================================\n\nfunction scoreHit(\n context: string,\n itemCode: string | null,\n _keyword: string,\n): number {\n const lower = context.toLowerCase();\n let score = 0;\n\n // Item-code boost\n if (itemCode && HIGH_SIGNAL_ITEMS[itemCode]) {\n score += HIGH_SIGNAL_ITEMS[itemCode].boost;\n }\n\n // Strategic neighbor words\n const neighbors = [\n \"initiative\",\n \"invest\",\n \"deploy\",\n \"implement\",\n \"launch\",\n \"reduce\",\n \"reduction\",\n \"eliminate\",\n \"transform\",\n \"strategy\",\n \"pilot\",\n \"acquired\",\n \"acquisition\",\n \"partnership\",\n \"committed\",\n \"priority\",\n \"accelerate\",\n \"expand\",\n ];\n for (const w of neighbors) if (lower.includes(w)) score += 1;\n\n // Mega-signal patterns\n const hasDollar = /\\$\\s*\\d+(?:\\.\\d+)?\\s*(?:million|billion|M|B)\\b/i.test(\n context,\n );\n const hasHeadcount =\n /\\b(headcount|severance|workforce reduction|layoff|reduction in force)/i\n .test(lower);\n const hasBackOffice =\n /\\b(back[\\s-]?office|administrative function|shared service)/i.test(lower);\n\n if (hasHeadcount && hasDollar) score += 8;\n if (hasBackOffice) score += 3;\n\n return score;\n}\n\nfunction scoreItem(item: ItemBlock): number {\n let score = HIGH_SIGNAL_ITEMS[item.item_code]?.boost || 0;\n if (/\\$\\s*\\d+(?:\\.\\d+)?\\s*(?:million|billion|M|B)\\b/i.test(item.content)) {\n score += 3;\n }\n if (\n /\\b(severance|layoff|reduction in force|exit charges?)\\b/i.test(\n item.content,\n )\n ) score += 4;\n return score;\n}\n\n// ============================================================================\n// Main handler\n// ============================================================================\n\nexport default async function handler(req: Request): Promise {\n const url = new URL(req.url);\n let ticker = (url.searchParams.get(\"ticker\") || \"\").toUpperCase().trim();\n\n if (!ticker && req.method === \"POST\") {\n try {\n const body = await req.json();\n ticker = (body.ticker || \"\").toUpperCase().trim();\n } catch { /* ignore */ }\n }\n\n if (!ticker) {\n return Response.json(\n {\n error: 'Pass ?ticker=AAPL or POST with {\"ticker\": \"AAPL\"}',\n usage: {\n get: \"GET /?ticker=AAPL\",\n post: 'POST / with {\"ticker\": \"AAPL\"}',\n },\n },\n { status: 400 },\n );\n }\n\n try {\n // Step 1: ticker -> CIK\n const tickersJson = await getTickers();\n let cik: string | null = null;\n let company: string | null = null;\n for (const entry of Object.values(tickersJson) as any[]) {\n if (entry.ticker.toUpperCase() === ticker) {\n cik = String(entry.cik_str).padStart(10, \"0\");\n company = entry.title;\n break;\n }\n }\n if (!cik) {\n return Response.json(\n { error: `Ticker ${ticker} not found in SEC EDGAR` },\n { status: 404 },\n );\n }\n\n // Step 2: recent 8-Ks from submissions API\n const subs = JSON.parse(\n await secFetch(`https://data.sec.gov/submissions/CIK${cik}.json`),\n );\n const recent = subs.filings.recent;\n\n type FilingInfo = { url: string; date: string; acc: string };\n const filings: FilingInfo[] = [];\n const cutoffDate = new Date(Date.now() - LOOKBACK_DAYS * 86400 * 1000)\n .toISOString().slice(0, 10);\n\n for (\n let i = 0; i < recent.form.length && filings.length < MAX_FILINGS; i++\n ) {\n if (recent.form[i] !== \"8-K\") continue;\n if (recent.filingDate[i] < cutoffDate) break;\n const acc = recent.accessionNumber[i].replace(/-/g, \"\");\n const doc = recent.primaryDocument[i];\n filings.push({\n url: `https://www.sec.gov/Archives/edgar/data/${\n parseInt(cik)\n }/${acc}/${doc}`,\n date: recent.filingDate[i],\n acc: recent.accessionNumber[i],\n });\n }\n\n if (filings.length === 0) {\n return Response.json(\n { error: `No 8-K found for ${ticker} in last ${LOOKBACK_DAYS} days` },\n { status: 404 },\n );\n }\n\n // Step 3: scan each 8-K\n type ItemHit = {\n item_code: string;\n item_title: string;\n content_preview: string;\n score: number;\n filing_date: string;\n filing_url: string;\n };\n type KeywordHit = {\n keyword: string;\n context: string;\n item_code: string | null;\n item_title: string | null;\n score: number;\n filing_date: string;\n filing_url: string;\n };\n\n const allItems: ItemHit[] = [];\n const allKwHits: KeywordHit[] = [];\n\n for (const filing of filings) {\n const html = await secFetch(filing.url);\n const text = stripHtml(html);\n\n const items = parseItems(text);\n\n for (const it of items) {\n allItems.push({\n item_code: it.item_code,\n item_title: it.item_title,\n content_preview: it.content.slice(0, 1500),\n score: scoreItem(it),\n filing_date: filing.date,\n filing_url: filing.url,\n });\n }\n\n for (const kw of KEYWORDS) {\n const pattern = new RegExp(\n \"\\\\b\" + kw.replace(/[.*+?^${}()|[\\]\\\\]/g, \"\\\\$&\") + \"\\\\b\",\n \"gi\",\n );\n let match;\n while ((match = pattern.exec(text)) !== null) {\n const charStart = Math.max(0, match.index - CONTEXT_CHARS);\n const charEnd = Math.min(\n text.length,\n match.index + kw.length + CONTEXT_CHARS,\n );\n const context = text.slice(charStart, charEnd).trim();\n const containingItem = findContainingItem(items, match.index);\n const score = scoreHit(\n context,\n containingItem?.item_code ?? null,\n kw,\n );\n\n allKwHits.push({\n keyword: kw,\n context,\n item_code: containingItem?.item_code ?? null,\n item_title: containingItem?.item_title ?? null,\n score,\n filing_date: filing.date,\n filing_url: filing.url,\n });\n }\n }\n }\n\n // Exact dedupe on keyword hits\n const seen = new Set();\n const uniqueKwHits: KeywordHit[] = [];\n for (const h of allKwHits) {\n const key = normalize(h.context);\n if (!seen.has(key)) {\n seen.add(key);\n uniqueKwHits.push(h);\n }\n }\n\n uniqueKwHits.sort((a, b) => b.score - a.score);\n allItems.sort((a, b) => b.score - a.score);\n\n const returnedHits = uniqueKwHits.slice(0, MAX_HITS_RETURNED);\n\n const kwCounts: Record = {};\n for (const h of uniqueKwHits) {\n kwCounts[h.keyword] = (kwCounts[h.keyword] || 0) + 1;\n }\n\n const itemCodeCounts: Record = {};\n for (const it of allItems) {\n itemCodeCounts[it.item_code] = (itemCodeCounts[it.item_code] || 0) + 1;\n }\n\n return Response.json({\n ticker,\n company,\n filings: filings.map((f) => ({ type: \"8-K\", date: f.date, url: f.url })),\n total_filings_scanned: filings.length,\n items_disclosed: allItems,\n item_code_counts: itemCodeCounts,\n total_keyword_hits_found: uniqueKwHits.length,\n total_keyword_hits_returned: returnedHits.length,\n keyword_counts: kwCounts,\n keyword_hits: returnedHits,\n });\n } catch (e: any) {\n return Response.json({ error: e.message }, { status: 500 });\n }\n}","http","createdAt",["D",1776976149731],"branchVersion","branchId","b0464b54-3f50-11f1-82b1-42b51c65c3df","id","b0453c32-3f50-11f1-82b1-42b51c65c3df","scraper-8k",["D",1776975121245],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","xx","imageUrl","828451e7-38d3-43a6-9686-9dd07ed10f76","pinnedFileId","pinPreview","pinMigrated","screenshotVersion","likeCount",0,"orgId","019dbb6e-9b96-734d-94c0-10d6676fe411","tursoName","tursoHostname","tursoMeasuredAt","pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_95":94,"_102":98},"org",{"_95":94,"_96":-5,"_97":98,"_18":55,"_99":100,"_101":25},"branch",{"_13":84,"_32":85,"_43":42,"_86":44,"_87":-5,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":45,"_93":94},"isOwned",6,"main","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","alexkrasiuk","handle","displayName","avatar","https://avatars.githubusercontent.com/u/274322468?v=4","tier","free","canRestrictVals","profileImageUrl","branchName","","adoptOrgId",{"_43":44,"_32":45,"_38":150,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":53,"_54":-5,"_18":55,"_56":-5,"_57":25,"_58":-5,"_59":-5,"_60":61,"_62":63,"_64":-5,"_65":-5,"_66":-5,"_67":-5,"_68":69,"_70":69,"_71":72,"_73":72,"_74":72,"_75":69,"_76":69,"_77":151,"_79":152,"_81":153,"_83":25},"tree",[145],"files",[145],"counts",{"_134":61,"_135":136,"_137":61,"_138":139,"_140":141},"contents",[126],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles",true,"grants","orgs","customDomains",{},{"_43":127,"_32":28,"_30":31,"_128":-5,"_41":42,"_40":14,"_129":25,"_38":130,"_13":131,"_33":37,"_35":36,"_24":25,"_132":-5,"_27":133},"1550f682-3f53-11f1-9404-42b51c65c3df","parentId","latest",["D",1776976149731],2,"label",[28],"pulls","branches",1,"forks","envVars",{"_142":61,"_144":61},"connections",{"_142":61,"_143":61},"count","unfilledCount","emptyCount",{"_43":127,"_32":28,"_30":31,"_128":-5,"_41":42,"_40":14,"_129":25,"_38":146,"_13":131,"_33":37,"_35":36,"_24":25,"_132":-5,"_147":148,"_27":149},["D",1776976149731],"children",[],[28],["D",1776975121245],{"_95":94,"_102":98},{"_95":94,"_96":-5,"_97":98,"_18":55,"_99":100,"_101":25},{"_13":84,"_32":85,"_43":42,"_86":44,"_87":-5,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":45,"_93":94},"ctx",{"_77":-5,"_123":187,"_188":189,"_190":191,"_192":25,"_193":25,"_194":25,"_195":121,"_196":-5,"_197":-7,"_198":-7,"_199":61,"_200":25,"_201":25,"_202":-5,"_203":25,"_204":25,"_205":25,"_206":-5},"ENV",{"_158":159,"_160":161,"_162":85,"_163":164,"_165":166,"_167":168,"_169":170,"_171":172,"_173":174,"_175":176,"_177":178,"_179":180,"_181":182,"_183":184,"_185":186},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","811aa22498c9b0b5a47e8d778de579cdb990f332","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]

Diff · v3