{title}

[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_272":273,"_274":275},"routes/_app.x.$handle.$name",{"_107":106,"_38":52,"_116":97,"_33":117,"_13":14,"_118":-5,"_11":119,"_120":121,"_122":123,"_124":125,"_126":127,"_26":-5,"_128":-5,"_129":-5,"_130":31,"_131":132,"_133":66,"_134":-5,"_135":-5,"_136":137},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":18,"_19":-5},"project",{"_50":51,"_38":52,"_45":53,"_54":-5,"_55":31,"_56":57,"_58":57,"_59":60,"_61":-5,"_19":62,"_63":64,"_65":66,"_67":-5,"_68":-5,"_69":70,"_71":72,"_73":74,"_75":76,"_77":78,"_79":-5,"_80":81,"_82":81,"_83":84,"_85":84,"_86":84,"_87":81,"_88":81,"_89":90,"_91":92,"_93":94,"_95":31},"version",25,"changes",[25],"aiChatInfo",{"_20":21,"_22":-5,"_23":-5,"_24":-5},"userId","aiChatMessageId","51294f4a-2f95-4be9-8c9b-ef5c9c04f671","aiChatThreadId","aiChatThreadUserId","aiChatThreadTitle",{"_26":27,"_28":29,"_30":31,"_32":31,"_33":34,"_35":34},"file",{"_36":37,"_38":39,"_40":41,"_42":44,"_45":46,"_47":14,"_48":49,"_30":31},"previous",{"_36":37,"_38":39,"_40":41,"_42":43},"deleted",false,"deletedHere","path","lib/gemini.ts","latestPath","fileId","82b9090e-a019-11f1-a74a-1607ee4eb77e","name","gemini.ts","fileType","script","content","// lib/gemini.ts\n// Gemini qualification/scoring layer. Reads GEMINI_API_KEY from the environment\n// at runtime — never hardcoded. Returns strict JSON we validate before trusting.\n\nimport { DiscoveryHit } from \"./sources.ts\";\n\nexport const MODEL = \"gemini-2.5-flash\";\n\n/** Opportunity as extracted by Gemini. All contact fields are nullable and only\n * ever populated from *public evidence* — never invented. */\nexport interface RawOpportunity {\n buyer_name?: string | null;\n company?: string | null;\n phone?: string | null;\n email?: string | null;\n /** What furniture product(s) and scope they need. */\n requirement: string;\n location?: string | null;\n source?: string | null;\n evidence_snippet?: string | null;\n /** 0-100 purchase intent. */\n intent_score?: number | null;\n /** Qualitative indicator of deal size where the evidence supports it. */\n opportunity_value?: { amount?: number; currency?: string } | null;\n confidence?: number | null;\n reject_reason?: string | null;\n}\n\nconst SYSTEM_PROMPT = `You are the IS-REAL AI Sales OS Market Hunter for IS-REAL Modern Furniture (Lagos, Nigeria).\nYour ONLY job is to find GENUINE, evidence-backed furniture buying opportunities in Nigeria from PUBLIC pages.\n\nTARGET PRODUCTS (from public buyer signal): sofas, wardrobes, beds, kitchen cabinets,\noffice furniture, hotel furniture, salon/spa furniture, construction/project furniture.\n\nRULES (non-negotiable):\n1. NEVER invent or guess a buyer name, company, phone number, email, budget or location.\n Anything you output must be explicitly stated in the evidence, or be null.\n2. Only include an item if there is a REAL purchasing intent (someone asking to buy,\n procure, supply, quote for, or furnish).\n3. Ignore: generic promo listings, \"cheapest furniture in town\", SEO spam, wholesale\n price lists that are not a buyer request, job ads, and unrelated content.\n4. Use evidence_url EXACTLY as given; do not fabricate one.\n5. Set intent_score 0 for anything you reject. Set reject_reason when skipping.\n6. opportunity_value is optional; set it ONLY when the page states a scale\n (number of items, project size, budget figure). Otherwise null.\n7. respond ONLY with a JSON object {\"opportunities\":[...]}`;\n\n/** Build the \"extract opportunities\" prompt from a batch of discovery hits. */\nexport function buildOpportunityPrompt(hits: DiscoveryHit[]): string {\n const blocks = hits\n .map(\n (h, i) =>\n `--- HIT ${i + 1} ---\\nURL: ${h.url}\\nTITLE: ${h.title}\\nTEXT: ${h.snippet}\\n`,\n )\n .join(\"\\n\");\n return `${SYSTEM_PROMPT}\\n\\nPUBLIC MARKET EVIDENCE:\\n${blocks}\\n\\n` +\n `Return JSON: {\"opportunities\":[{\"buyer_name\":..,\"company\":..,\"phone\":..,\"email\":..,\"requirement\":..,\"location\":..,\"intent_score\":0-100,\"opportunity_value\":null|{amount,currency},\"confidence\":0-1,\"evidence_snippet\":..,\"reject_reason\":null|..}]}`;\n}\n\n/** Call Gemini with a JSON-only response contract. */\nexport async function geminiJson(prompt: string): Promise {\n const key = Deno.env.get(\"GEMINI_API_KEY\");\n if (!key) {\n throw new Error(\"GEMINI_API_KEY is not configured\");\n }\n const url =\n `https://generativelanguage.googleapis.com/v1beta/models/${MODEL}:generateContent?key=` +\n encodeURIComponent(key);\n const body = {\n contents: [{ parts: [{ text: prompt }] }],\n generationConfig: {\n responseMimeType: \"application/json\",\n responseSchema: {\n type: \"OBJECT\",\n properties: {\n opportunities: {\n type: \"ARRAY\",\n items: {\n type: \"OBJECT\",\n properties: {\n buyer_name: { type: \"STRING\", nullable: true },\n company: { type: \"STRING\", nullable: true },\n phone: { type: \"STRING\", nullable: true },\n email: { type: \"STRING\", nullable: true },\n requirement: { type: \"STRING\", nullable: true },\n location: { type: \"STRING\", nullable: true },\n intent_score: { type: \"INTEGER\", nullable: true },\n opportunity_value: {\n type: \"OBJECT\",\n nullable: true,\n properties: {\n amount: { type: \"NUMBER\", nullable: true },\n currency: { type: \"STRING\", nullable: true },\n },\n },\n confidence: { type: \"NUMBER\", nullable: true },\n evidence_snippet: { type: \"STRING\", nullable: true },\n reject_reason: { type: \"STRING\", nullable: true },\n },\n },\n },\n },\n },\n },\n };\n const res = await fetch(url, {\n method: \"POST\",\n headers: { \"content-type\": \"application/json\" },\n body: JSON.stringify(body),\n });\n if (!res.ok) {\n const detail = await res.text().catch(() => \"\");\n throw new Error(`Gemini ${res.status}: ${detail.slice(0, 400)}`);\n }\n const json = await res.json();\n const text = json?.candidates?.[0]?.content?.parts?.[0]?.text ?? \"\";\n return parseJsonish(text);\n}\n\n/** Robustly parse JSON out of an LLM response (handles code fences). */\nexport function parseJsonish(text: string): T {\n const cleaned = text.replace(/```(?:json)?/gi, \"\").trim();\n return JSON.parse(cleaned) as T;\n}\n","// lib/gemini.ts\n// Gemini qualification/scoring layer. Reads GEMINI_API_KEY from the environment\n// at runtime — never hardcoded. Returns strict JSON we validate before trusting.\n\nimport { DiscoveryHit } from \"./sources.ts\";\n\nexport const MODEL = \"gemini-3.6-flash\";\n\n/** Opportunity as extracted by Gemini. All contact fields are nullable and only\n * ever populated from *public evidence* — never invented. */\nexport interface RawOpportunity {\n buyer_name?: string | null;\n company?: string | null;\n phone?: string | null;\n email?: string | null;\n /** What furniture product(s) and scope they need. */\n requirement: string;\n location?: string | null;\n source?: string | null;\n evidence_snippet?: string | null;\n /** 0-100 purchase intent. */\n intent_score?: number | null;\n /** Qualitative indicator of deal size where the evidence supports it. */\n opportunity_value?: { amount?: number; currency?: string } | null;\n confidence?: number | null;\n reject_reason?: string | null;\n}\n\nconst SYSTEM_PROMPT = `You are the IS-REAL AI Sales OS Market Hunter for IS-REAL Modern Furniture (Lagos, Nigeria).\nYour ONLY job is to find GENUINE, evidence-backed furniture buying opportunities in Nigeria from PUBLIC pages.\n\nTARGET PRODUCTS (from public buyer signal): sofas, wardrobes, beds, kitchen cabinets,\noffice furniture, hotel furniture, salon/spa furniture, construction/project furniture.\n\nRULES (non-negotiable):\n1. NEVER invent or guess a buyer name, company, phone number, email, budget or location.\n Anything you output must be explicitly stated in the evidence, or be null.\n2. Only include an item if there is a REAL purchasing intent (someone asking to buy,\n procure, supply, quote for, or furnish).\n3. Ignore: generic promo listings, \"cheapest furniture in town\", SEO spam, wholesale\n price lists that are not a buyer request, job ads, and unrelated content.\n4. Use evidence_url EXACTLY as given; do not fabricate one.\n5. Set intent_score 0 for anything you reject. Set reject_reason when skipping.\n6. opportunity_value is optional; set it ONLY when the page states a scale\n (number of items, project size, budget figure). Otherwise null.\n7. respond ONLY with a JSON object {\"opportunities\":[...]}`;\n\n/** Build the \"extract opportunities\" prompt from a batch of discovery hits. */\nexport function buildOpportunityPrompt(hits: DiscoveryHit[]): string {\n const blocks = hits\n .map(\n (h, i) =>\n `--- HIT ${i + 1} ---\\nURL: ${h.url}\\nTITLE: ${h.title}\\nTEXT: ${h.snippet}\\n`,\n )\n .join(\"\\n\");\n return `${SYSTEM_PROMPT}\\n\\nPUBLIC MARKET EVIDENCE:\\n${blocks}\\n\\n` +\n `Return JSON: {\"opportunities\":[{\"buyer_name\":..,\"company\":..,\"phone\":..,\"email\":..,\"requirement\":..,\"location\":..,\"intent_score\":0-100,\"opportunity_value\":null|{amount,currency},\"confidence\":0-1,\"evidence_snippet\":..,\"reject_reason\":null|..}]}`;\n}\n\n/** Call Gemini with a JSON-only response contract. */\nexport async function geminiJson(prompt: string): Promise {\n const key = Deno.env.get(\"GEMINI_API_KEY\");\n if (!key) {\n throw new Error(\"GEMINI_API_KEY is not configured\");\n }\n const url =\n `https://generativelanguage.googleapis.com/v1beta/models/${MODEL}:generateContent?key=` +\n encodeURIComponent(key);\n const body = {\n contents: [{ parts: [{ text: prompt }] }],\n generationConfig: {\n responseMimeType: \"application/json\",\n responseSchema: {\n type: \"OBJECT\",\n properties: {\n opportunities: {\n type: \"ARRAY\",\n items: {\n type: \"OBJECT\",\n properties: {\n buyer_name: { type: \"STRING\", nullable: true },\n company: { type: \"STRING\", nullable: true },\n phone: { type: \"STRING\", nullable: true },\n email: { type: \"STRING\", nullable: true },\n requirement: { type: \"STRING\", nullable: true },\n location: { type: \"STRING\", nullable: true },\n intent_score: { type: \"INTEGER\", nullable: true },\n opportunity_value: {\n type: \"OBJECT\",\n nullable: true,\n properties: {\n amount: { type: \"NUMBER\", nullable: true },\n currency: { type: \"STRING\", nullable: true },\n },\n },\n confidence: { type: \"NUMBER\", nullable: true },\n evidence_snippet: { type: \"STRING\", nullable: true },\n reject_reason: { type: \"STRING\", nullable: true },\n },\n },\n },\n },\n },\n },\n };\n const res = await fetch(url, {\n method: \"POST\",\n headers: { \"content-type\": \"application/json\" },\n body: JSON.stringify(body),\n });\n if (!res.ok) {\n const detail = await res.text().catch(() => \"\");\n throw new Error(`Gemini ${res.status}: ${detail.slice(0, 400)}`);\n }\n const json = await res.json();\n const text = json?.candidates?.[0]?.content?.parts?.[0]?.text ?? \"\";\n return parseJsonish(text);\n}\n\n/** Robustly parse JSON out of an LLM response (handles code fences). */\nexport function parseJsonish(text: string): T {\n const cleaned = text.replace(/```(?:json)?/gi, \"\").trim();\n return JSON.parse(cleaned) as T;\n}\n","createdAt",["D",1787619033037],"branchVersion","branchId","ac292285-e807-44ad-b932-d224d7a2bc0b","id","062aa035-2622-42e9-b197-d325a6afd62b","is-real-ai-sales-os",["D",1787435303615],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","IS-REAL AI client acquisition and sales automation OS","imageUrl","7f4d7826-53c8-4b3e-9b4f-1e07fc1853e4","pinnedFileId","01a02b72-4343-720e-b974-db3653f8238f","pinPreview",true,"pinMigrated","screenshotVersion","likeCount",0,"orgId","01a02b69-5d30-761f-af3c-cfbcd92a2566","tursoName","062aa035262242e9b197d325a6afd62b","tursoHostname","062aa035262242e9b197d325a6afd62b-val-town.aws-us-east-2.turso.io","tursoMeasuredAt",["D",1789521640046],"pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_107":106,"_115":111},"org",{"_107":106,"_108":109,"_110":111,"_19":62,"_112":113,"_114":31},"branch",{"_13":96,"_38":97,"_50":49,"_98":51,"_99":-5,"_100":-5,"_101":-5,"_102":-5,"_103":-5,"_104":52,"_105":106},"isOwned",112,"main","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","isrealmodernfurniture","handle","displayName","IS-REAL Modern furniture","avatar","https://lh3.googleusercontent.com/a/ACg8ocLoG7wn933Ub4ABjB-fo6OOcMHznhw2GQGgVPJAJXagSLjgyeo7=s96-c","tier","free","canRestrictVals","profileImageUrl","branchName","","adoptOrgId",{"_50":51,"_38":52,"_45":267,"_54":-5,"_55":31,"_56":57,"_58":57,"_59":60,"_61":-5,"_19":62,"_63":64,"_65":66,"_67":-5,"_68":-5,"_69":70,"_71":72,"_73":74,"_75":76,"_77":268,"_79":-5,"_80":81,"_82":81,"_83":84,"_85":84,"_86":84,"_87":81,"_88":81,"_89":269,"_91":270,"_93":271,"_95":31},"tree",[208,210,214,215,216,217,218],"files",[208,209,210,211,212,213,214,215,216,217,218],"counts",{"_197":70,"_198":152,"_199":70,"_200":201,"_202":203},"contents",[138,139,140,141,142,143,144],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},{"_50":192,"_38":193,"_36":194,"_148":-5,"_48":49,"_47":152,"_150":66,"_45":195,"_13":70,"_40":190,"_42":-5,"_30":31,"_155":-5,"_33":196},{"_50":186,"_38":187,"_36":188,"_148":-5,"_48":49,"_47":161,"_150":66,"_45":189,"_13":70,"_40":190,"_42":-5,"_30":31,"_155":-5,"_33":191},{"_50":180,"_38":181,"_36":182,"_148":-5,"_48":49,"_47":152,"_150":66,"_45":183,"_13":70,"_40":26,"_42":184,"_30":31,"_155":-5,"_33":185},{"_50":173,"_38":174,"_36":175,"_148":-5,"_48":49,"_47":176,"_150":66,"_45":177,"_13":152,"_40":26,"_42":178,"_30":31,"_155":-5,"_33":179},{"_50":166,"_38":167,"_36":168,"_148":-5,"_48":49,"_47":169,"_150":31,"_45":170,"_13":70,"_40":41,"_42":171,"_30":31,"_155":-5,"_33":172},{"_50":157,"_38":158,"_36":64,"_148":-5,"_48":49,"_47":159,"_150":31,"_45":160,"_13":161,"_40":162,"_42":163,"_30":31,"_155":164,"_33":165},{"_50":145,"_38":146,"_36":147,"_148":-5,"_48":49,"_47":149,"_150":31,"_45":151,"_13":152,"_40":153,"_42":154,"_30":31,"_155":-5,"_33":156},"8390c3b5-2799-43b0-8370-0eb3d812b47f","market-hunter.ts","68d53af0-9e73-11f1-9179-1607ee4eb77e","parentId",8,"latest",["D",1787616751409],1,"interval","// market-hunter.ts\n// Scheduled Market Hunt worker (fileType: interval). Runs automatically on the\n// cron schedule. Each execution discovers public furniture-buying evidence in\n// Nigeria, qualifies it with Gemini, stores qualified leads, and logs a clear\n// result. Invoking this file directly (e.g. via run_file) triggers one hunt.\n\nimport { runHunt } from \"./lib/hunter.ts\";\n\nexport default async function (interval: Interval) {\n console.log(\n `[market-hunter] scheduled run started. lastRunAt=${interval.lastRunAt ? interval.lastRunAt.toISOString() : \"(first run)\"}`,\n );\n\n const result = await runHunt({ source: \"cron\" });\n\n // Clear, structured result for the schedule trigger and for logs/observability.\n console.log(\n JSON.stringify({\n market_hunter: \"IS-REAL AI Sales OS\",\n run: result.status,\n ok: result.ok,\n huntId: result.huntId,\n triggeredBy: result.triggeredBy,\n startedAt: result.startedAt,\n finishedAt: result.finishedAt,\n queriesRun: result.queriesRun,\n hitsSeen: result.hitsSeen,\n qualifiedCandidates: result.candidates,\n newLeads: result.newLeads,\n duplicatesDeduped: result.duplicates,\n rejected: result.rejected,\n geminiConfigured: Boolean(Deno.env.get(\"GEMINI_API_KEY\")),\n error: result.error ?? null,\n }),\n );\n\n return result;\n}\n","label",[146],"53194ea2-5b58-4f96-a287-f81a472dc1a6","index.ts",23,["D",1787618242254],3,"http","/** @jsxImportSource npm:hono@4/jsx */\n// index.ts\n// IS-REAL AI Sales OS — Market Hunter HTTP dashboard + API.\n// Serves a live ops dashboard (real SQLite data) and REST endpoints, including a\n// manual \"Run Hunt Now\" trigger. Server-rendered with hono/jsx (no client SPA).\n\nimport { Hono } from \"npm:hono\";\nimport { parseVal } from \"https://esm.town/v/std/utils/index.ts\";\nimport {\n BUSINESS_WHATSAPP,\n businessWhatsApp,\n getLead,\n listHunts,\n listLeads,\n runHunt,\n setLeadStatus,\n stats,\n WEBSITE,\n} from \"./lib/hunter.ts\";\nimport type { LeadRow } from \"./lib/hunter.ts\";\n\nconst app = new Hono();\n\nconst VALID_STATUSES = [\"new\", \"contacted\", \"quoted\", \"closed\", \"rejected\"] as const;\n\nconst fmt = (n: number | null, cur = \"NGN\") =>\n n == null ? \"—\" : `${cur} ${Number(n).toLocaleString(\"en-NG\")}`;\n\nfunction Badge({ children, tone }: { children: unknown; tone?: string }) {\n const color =\n tone === \"error\"\n ? \"#e74c3c\"\n : tone === \"ok\"\n ? \"#27ae60\"\n : tone === \"warn\"\n ? \"#e67e22\"\n : \"#3498db\";\n return (\n \n {children}\n \n );\n}\n\nfunction Layout({ title, children }: { title: string; children: unknown }) {\n return (\n \n \n \n \n {title}\n

Diff · v25