mcp-scraper 0.4.5 → 0.4.12

package/dist/bin/api-server.cjs

@@ -3617,6 +3617,7 @@ __export(db_exports, {
   getLedger: () => getLedger,
   getMemoryPlan: () => getMemoryPlan,
   getMemoryProvisioned: () => getMemoryProvisioned,
+  getPageSnapshot: () => getPageSnapshot,
   getRefresh: () => getRefresh,
   getUserByApiKey: () => getUserByApiKey,
   getUserByEmail: () => getUserByEmail,
@@ -3672,6 +3673,7 @@ __export(db_exports, {
   startHarvestAttempt: () => startHarvestAttempt,
   stripeEventAlreadyProcessed: () => stripeEventAlreadyProcessed,
   updateKpoJobState: () => updateKpoJobState,
+  upsertPageSnapshot: () => upsertPageSnapshot,
   verifyPassword: () => verifyPassword
 });
 function getDb() {
@@ -4109,6 +4111,20 @@ async function migrate() {
     } catch {
     }
   }
+  await db.execute(`
+    CREATE TABLE IF NOT EXISTS page_snapshots (
+      user_id       INTEGER NOT NULL REFERENCES users(id),
+      url           TEXT    NOT NULL,
+      content_hash  TEXT    NOT NULL,
+      content       TEXT    NOT NULL,
+      title         TEXT,
+      content_bytes INTEGER NOT NULL,
+      truncated     INTEGER NOT NULL DEFAULT 0,
+      checked_at    TEXT    NOT NULL DEFAULT (datetime('now')),
+      PRIMARY KEY (user_id, url)
+    )
+  `);
+  await db.execute(`CREATE INDEX IF NOT EXISTS page_snapshots_user_checked_at ON page_snapshots(user_id, checked_at DESC)`);
 }
 async function registerClient(clientId, redirectUris, name) {
   await getDb().execute({
@@ -4539,6 +4555,38 @@ async function listJobs(userId) {
   const res = await getDb().execute({ sql: "SELECT * FROM jobs WHERE user_id = ? ORDER BY created_at DESC LIMIT 50", args: [userId] });
   return res.rows.map((r) => deserialize(rowToRawJob(r)));
 }
+async function getPageSnapshot(userId, url) {
+  const res = await getDb().execute({
+    sql: "SELECT content_hash, content, title, content_bytes, truncated, checked_at FROM page_snapshots WHERE user_id = ? AND url = ?",
+    args: [userId, url]
+  });
+  const row = res.rows[0];
+  if (!row) return null;
+  return {
+    contentHash: String(row.content_hash),
+    content: String(row.content),
+    title: row.title == null ? null : String(row.title),
+    contentBytes: Number(row.content_bytes),
+    truncated: Number(row.truncated) === 1,
+    checkedAt: String(row.checked_at)
+  };
+}
+async function upsertPageSnapshot(userId, url, snapshot) {
+  await getDb().execute({
+    sql: `
+      INSERT INTO page_snapshots (user_id, url, content_hash, content, title, content_bytes, truncated, checked_at)
+      VALUES (?, ?, ?, ?, ?, ?, ?, datetime('now'))
+      ON CONFLICT(user_id, url) DO UPDATE SET
+        content_hash  = excluded.content_hash,
+        content       = excluded.content,
+        title         = excluded.title,
+        content_bytes = excluded.content_bytes,
+        truncated     = excluded.truncated,
+        checked_at    = excluded.checked_at
+    `,
+    args: [userId, url, snapshot.contentHash, snapshot.content, snapshot.title, snapshot.contentBytes, snapshot.truncated ? 1 : 0]
+  });
+}
 function rowToRequestEvent(row) {
   const rawResult = row.result != null ? String(row.result) : null;
   return {
@@ -5293,18 +5341,25 @@ function kernelCostUsd(ms, headful) {
   return sec * (headful ? KERNEL_HEADFUL_USD_PER_SEC : KERNEL_HEADLESS_USD_PER_SEC);
 }
 function vendorCostUsd(vendor, units) {
-  if (vendor === "fal_wizper") return Math.max(0, units) * FAL_WIZPER_USD_PER_AUDIO_MIN;
+  if (vendor === "fal_wizper") {
+    const estimatedComputeSec = Math.max(0, units) / FAL_WIZPER_WALLCLOCK_TO_COMPUTE_SEC_FACTOR;
+    return estimatedComputeSec * FAL_WIZPER_USD_PER_COMPUTE_SEC;
+  }
   if (vendor === "deepinfra_qwen") return Math.max(0, units) / 1e3 * DEEPINFRA_QWEN_USD_PER_1K_OUT_TOKENS;
+  if (vendor === "openrouter") return Math.max(0, units);
+  if (vendor === "mcp_memory_video") return Math.max(0, units);
+  if (vendor === "mcp_memory_ai") return Math.max(0, units);
   return 0;
 }
-var KERNEL_HEADLESS_USD_PER_SEC, KERNEL_HEADFUL_USD_PER_SEC, HEADLESS_OPS, FAL_WIZPER_USD_PER_AUDIO_MIN, DEEPINFRA_QWEN_USD_PER_1K_OUT_TOKENS;
+var KERNEL_HEADLESS_USD_PER_SEC, KERNEL_HEADFUL_USD_PER_SEC, HEADLESS_OPS, FAL_WIZPER_USD_PER_COMPUTE_SEC, FAL_WIZPER_WALLCLOCK_TO_COMPUTE_SEC_FACTOR, DEEPINFRA_QWEN_USD_PER_1K_OUT_TOKENS;
 var init_cost_rates = __esm({
   "src/api/cost-rates.ts"() {
     "use strict";
     KERNEL_HEADLESS_USD_PER_SEC = 166667e-10;
     KERNEL_HEADFUL_USD_PER_SEC = 1333336e-10;
-    HEADLESS_OPS = /* @__PURE__ */ new Set(["harvest", "maps_search", "yt_channel", "fb_search", "fb_ad", "instagram"]);
-    FAL_WIZPER_USD_PER_AUDIO_MIN = 5e-3;
+    HEADLESS_OPS = /* @__PURE__ */ new Set(["serp", "maps_search", "yt_channel", "yt_transcription", "fb_search", "fb_ad", "instagram"]);
+    FAL_WIZPER_USD_PER_COMPUTE_SEC = 111e-5;
+    FAL_WIZPER_WALLCLOCK_TO_COMPUTE_SEC_FACTOR = 4;
     DEEPINFRA_QWEN_USD_PER_1K_OUT_TOKENS = 4e-4;
   }
 });
@@ -5345,6 +5400,10 @@ async function migrateCostTelemetry() {
     await db.execute(`ALTER TABLE kernel_session_log ADD COLUMN proxy_type TEXT`);
   } catch {
   }
+  try {
+    await db.execute(`ALTER TABLE kernel_session_log ADD COLUMN method TEXT`);
+  } catch {
+  }
   await db.execute(`
     CREATE TABLE IF NOT EXISTS vendor_usage_log (
       id           TEXT PRIMARY KEY,
@@ -5363,6 +5422,10 @@ async function migrateCostTelemetry() {
   await db.execute(`CREATE INDEX IF NOT EXISTS vendor_usage_log_op ON vendor_usage_log(op)`);
   await db.execute(`CREATE INDEX IF NOT EXISTS vendor_usage_log_probe ON vendor_usage_log(probe_run_id)`);
   await db.execute(`CREATE INDEX IF NOT EXISTS vendor_usage_log_created ON vendor_usage_log(created_at)`);
+  try {
+    await db.execute(`ALTER TABLE vendor_usage_log ADD COLUMN method TEXT`);
+  } catch {
+  }
   await db.execute(`
     CREATE TABLE IF NOT EXISTS cost_probe_runs (
       id                            TEXT PRIMARY KEY,
@@ -5409,8 +5472,8 @@ async function recordKernelSession(r) {
     const db = getDb();
     await db.execute({
       sql: `INSERT INTO kernel_session_log
-        (id, kernel_session_id, op, source, probe_run_id, user_id, stealth, headless_sent, proxy_used, proxy_source, proxy_type, fallback, opened_at, closed_at, duration_ms, est_cost_usd_headless, est_cost_usd_headful, error)
-        VALUES (?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)`,
+        (id, kernel_session_id, op, source, probe_run_id, user_id, stealth, headless_sent, proxy_used, proxy_source, proxy_type, fallback, opened_at, closed_at, duration_ms, est_cost_usd_headless, est_cost_usd_headful, error, method)
+        VALUES (?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)`,
       args: [
         (0, import_node_crypto2.randomUUID)(),
         r.kernelSessionId ?? null,
@@ -5429,7 +5492,8 @@ async function recordKernelSession(r) {
         durationMs,
         kernelCostUsd(durationMs, false),
         kernelCostUsd(durationMs, true),
-        r.error ?? null
+        r.error ?? null,
+        ctx?.subOp ?? null
       ]
     });
   } catch (err) {
@@ -5443,8 +5507,8 @@ async function recordVendorUsage(r) {
     const db = getDb();
     await db.execute({
       sql: `INSERT INTO vendor_usage_log
-        (id, op, probe_run_id, user_id, vendor, model, units, unit_type, est_cost_usd, error)
-        VALUES (?,?,?,?,?,?,?,?,?,?)`,
+        (id, op, probe_run_id, user_id, vendor, model, units, unit_type, est_cost_usd, error, method)
+        VALUES (?,?,?,?,?,?,?,?,?,?,?)`,
       args: [
         (0, import_node_crypto2.randomUUID)(),
         ctx?.op ?? null,
@@ -5455,7 +5519,8 @@ async function recordVendorUsage(r) {
         r.units,
         r.unitType,
         vendorCostUsd(r.vendor, r.units),
-        r.error ?? null
+        r.error ?? null,
+        ctx?.subOp ?? null
       ]
     });
   } catch (err) {
@@ -7357,6 +7422,7 @@ var init_llm_parse_with_retry = __esm({
     DeepInfraLlmClient = class {
       apiKey;
       model;
+      totalCostUsd = 0;
       constructor(apiKey, model = "Qwen/Qwen3.6-35B-A3B") {
         this.apiKey = apiKey;
         this.model = model;
@@ -7389,6 +7455,7 @@ ${raw}`
     OpenRouterLlmClient = class {
       apiKey;
       model;
+      totalCostUsd = 0;
       constructor(apiKey, model = "qwen/qwen3-235b-a22b") {
         this.apiKey = apiKey;
         this.model = model;
@@ -7401,6 +7468,9 @@ ${raw}`
         });
         if (!resp.ok) throw new Error(`OpenRouter error ${resp.status}: ${await resp.text()}`);
         const data = await resp.json();
+        const cost = data.usage?.cost ?? 0;
+        this.totalCostUsd += cost;
+        void recordVendorUsage({ vendor: "openrouter", model: this.model, units: cost, unitType: "usd" });
         return data.choices[0].message.content;
       }
       async completeJson(prompt) {
@@ -7426,6 +7496,9 @@ ${raw}`
       secondary;
       lastServedBy = null;
       lastPrimaryError = null;
+      get totalCostUsd() {
+        return this.primary.totalCostUsd + this.secondary.totalCostUsd;
+      }
       async complete(prompt) {
         try {
           const result = await this.primary.complete(prompt);
@@ -7880,6 +7953,390 @@ var init_schemas = __esm({
   }
 });
 
+// src/api/rates.ts
+function browserActiveCostMc(activeMs) {
+  return Math.round(activeMs * MC_PER_BROWSER_MS);
+}
+function concurrencySlotBillingInfo() {
+  const billingUrl = process.env.CONCURRENCY_BILLING_URL ?? "https://mcpscraper.dev/billing";
+  return {
+    product: "Extra concurrency slot",
+    price_label: CONCURRENCY_SLOT_PRICE_LABEL,
+    unit_amount_usd: CONCURRENCY_SLOT_PRICE_USD,
+    currency: CONCURRENCY_SLOT_PRICE_CURRENCY,
+    interval: CONCURRENCY_SLOT_PRICE_INTERVAL,
+    billing_url: billingUrl,
+    terminal_command: CONCURRENCY_SLOT_TERMINAL_COMMAND,
+    terminal_command_with_api_key_env: `MCP_SCRAPER_API_KEY=sk_live_your_key ${CONCURRENCY_SLOT_TERMINAL_COMMAND}`
+  };
+}
+function mcToCredits(mc) {
+  return mc / MC_PER_CREDIT;
+}
+function insufficientBalanceResponse(balanceMc, requiredMc) {
+  const topupUrl = process.env.TOPUP_URL ?? "https://mcpscraper.dev/billing";
+  const balanceCredits = mcToCredits(balanceMc);
+  const requiredCredits = mcToCredits(requiredMc);
+  return {
+    error: "insufficient_balance",
+    error_code: "insufficient_balance",
+    message: `Insufficient credits. Balance: ${balanceCredits} credits. This call requires ${requiredCredits} credits. Top up at ${topupUrl}`,
+    balance_credits: balanceCredits,
+    required_credits: requiredCredits,
+    topup_url: topupUrl
+  };
+}
+var MC_COSTS, MC_PER_BROWSER_MS, BROWSER_OPEN_MIN_BALANCE_MC, MC_PER_CREDIT, CREDIT_COST_CATALOG, CONCURRENCY_PRICE_ID, SUBSCRIPTION_TIERS, SUBSCRIPTION_TIER_BY_KEY, CONCURRENCY_SLOT_PRICE_USD, CONCURRENCY_SLOT_PRICE_CURRENCY, CONCURRENCY_SLOT_PRICE_INTERVAL, CONCURRENCY_SLOT_PRICE_LABEL, CONCURRENCY_SLOT_TERMINAL_COMMAND, FREE_SIGNUP_MC, MEMORY_FREE_QUOTA_BYTES, MEMORY_PLANS, MEMORY_PLAN_QUOTA, MEMORY_MARGIN_MULTIPLE, MEMORY_PRO_COST_BUDGET_USD, SCHEDULING_PLANS, LedgerOperation, MEMORY_AI_MARGIN_MULTIPLE, SITE_AUDIT_LLM_MARGIN_MULTIPLE, VIDEO_ANALYSIS_LLM_MARGIN_MULTIPLE, MC_PER_USD;
+var init_rates = __esm({
+  "src/api/rates.ts"() {
+    "use strict";
+    MC_COSTS = {
+      serp: 100,
+      paa: 200,
+      paa_base: 1e3,
+      page_scrape: 100,
+      url_map: 500,
+      yt_channel: 200,
+      yt_transcription: 200,
+      fb_ad: 200,
+      maps_search: 500,
+      maps_place: 6e3,
+      maps_review: 100,
+      fb_search: 600,
+      fb_transcribe: 200,
+      google_ads_search: 600,
+      google_ads_intel: 200,
+      google_ads_transcribe: 200,
+      instagram_profile: 400,
+      instagram_media: 400,
+      instagram_transcribe: 50,
+      browser_minute: 12e3,
+      reddit_thread: 3e3,
+      reddit_comment: 200,
+      video_analysis: 666667,
+      trustpilot_reviews: 500,
+      trustpilot_review: 100,
+      g2_reviews: 500,
+      g2_review: 150,
+      diff_page: 100
+    };
+    MC_PER_BROWSER_MS = MC_COSTS.browser_minute / 6e4;
+    BROWSER_OPEN_MIN_BALANCE_MC = 1e3;
+    MC_PER_CREDIT = 1e3;
+    CREDIT_COST_CATALOG = [
+      {
+        key: "serp",
+        label: "SERP search",
+        aliases: ["search_serp", "serp", "google search", "organic results"],
+        credits: mcToCredits(MC_COSTS.serp),
+        unit: "per search",
+        notes: "Returns AI Overview, PAA snippet, videos, forums, and local pack."
+      },
+      {
+        key: "paa",
+        label: "PAA harvest",
+        aliases: ["harvest_paa", "paa", "people also ask", "questions"],
+        credits: mcToCredits(MC_COSTS.paa),
+        unit: `per question (+${mcToCredits(MC_COSTS.paa_base)} credit base)`,
+        notes: `Full SERP feature extraction. Billed ${mcToCredits(MC_COSTS.paa_base)} credit base + ${mcToCredits(MC_COSTS.paa)} per question actually returned (unused estimate refunded).`
+      },
+      {
+        key: "page_scrape",
+        label: "Page crawl / extract",
+        aliases: ["extract_url", "extract_site", "page scrape", "url scrape", "single page", "site crawl"],
+        credits: mcToCredits(MC_COSTS.page_scrape),
+        unit: "per page",
+        notes: "Applies to both single-URL extraction and per-page site crawls."
+      },
+      {
+        key: "url_map",
+        label: "Site URL mapping",
+        aliases: ["map_site_urls", "url map", "site map", "crawl urls"],
+        credits: mcToCredits(MC_COSTS.url_map),
+        unit: "per mapping operation",
+        notes: "Flat rate for the full /map-urls call regardless of URL count discovered."
+      },
+      {
+        key: "yt_channel",
+        label: "YouTube search / channel harvest",
+        aliases: ["youtube_harvest", "youtube search", "youtube channel", "yt_channel"],
+        credits: mcToCredits(MC_COSTS.yt_channel),
+        unit: "per call"
+      },
+      {
+        key: "yt_transcription",
+        label: "YouTube transcription",
+        aliases: ["youtube_transcribe", "youtube transcript", "transcription", "yt_transcription"],
+        credits: mcToCredits(MC_COSTS.yt_transcription),
+        unit: "per minute",
+        notes: "A 5-minute hold is taken, then reconciled to actual video duration."
+      },
+      {
+        key: "fb_ad",
+        label: "Facebook search / ad lookup",
+        aliases: ["facebook_page_intel", "facebook_ad_search", "facebook_ad", "facebook ads", "fb ads"],
+        credits: mcToCredits(MC_COSTS.fb_ad),
+        unit: "per call"
+      },
+      {
+        key: "maps_search",
+        label: "Maps business search",
+        aliases: ["maps_search", "google maps search", "gmb search", "gbp search", "business profiles"],
+        credits: mcToCredits(MC_COSTS.maps_search),
+        unit: "per search",
+        notes: "Returns up to 50 Google Maps business/profile candidates. Use maps_place_intel to hydrate selected businesses."
+      },
+      {
+        key: "maps_place",
+        label: "Maps business lookup",
+        aliases: ["maps_place_intel", "google maps", "maps place", "place intel"],
+        credits: mcToCredits(MC_COSTS.maps_place),
+        unit: "per business",
+        notes: "Base lookup. Reviews billed separately per card at maps_review rate."
+      },
+      {
+        key: "maps_review",
+        label: "Maps review",
+        aliases: ["maps_reviews", "google reviews", "review cards", "reviews"],
+        credits: mcToCredits(MC_COSTS.maps_review),
+        unit: "per review card",
+        notes: "Charged after extraction when includeReviews is true."
+      },
+      {
+        key: "fb_search",
+        label: "Facebook ad library search",
+        aliases: ["facebook_search", "fb_search", "fb ad search"],
+        credits: mcToCredits(MC_COSTS.fb_search),
+        unit: "per search",
+        notes: "Browser automation to search Facebook Ads Library by keyword."
+      },
+      {
+        key: "google_ads_search",
+        label: "Google Ads Transparency search",
+        aliases: ["google_ads_search", "google ads search", "ads transparency search"],
+        credits: mcToCredits(MC_COSTS.google_ads_search),
+        unit: "per search",
+        notes: "Browser automation to find advertisers in Google Ads Transparency Center by domain or name."
+      },
+      {
+        key: "google_ads_intel",
+        label: "Google Ads Transparency advertiser intel",
+        aliases: ["google_ads_page_intel", "google ads intel", "ads transparency intel"],
+        credits: mcToCredits(MC_COSTS.google_ads_intel),
+        unit: "per call",
+        notes: "Lists and hydrates an advertiser's creatives with image URLs and video references."
+      },
+      {
+        key: "google_ads_transcribe",
+        label: "Google ad video transcription",
+        aliases: ["google_ads_transcribe", "google ad transcript"],
+        credits: mcToCredits(MC_COSTS.google_ads_transcribe),
+        unit: "per minute",
+        notes: "A hold is taken, then reconciled to actual video duration."
+      },
+      {
+        key: "fb_transcribe",
+        label: "Facebook video / ad transcription",
+        aliases: ["facebook_transcribe", "facebook_video_transcribe", "fb_transcribe", "fb ad transcript"],
+        credits: mcToCredits(MC_COSTS.fb_transcribe),
+        unit: "per minute",
+        notes: "A hold is taken, then reconciled to actual video duration."
+      },
+      {
+        key: "instagram_profile",
+        label: "Instagram profile content discovery",
+        aliases: ["instagram_profile_content", "instagram profile", "instagram content list", "ig profile"],
+        credits: mcToCredits(MC_COSTS.instagram_profile),
+        unit: "per profile scan",
+        notes: "Browser extraction of public Instagram profile grid links. Complete history may require a logged-in profile."
+      },
+      {
+        key: "instagram_media",
+        label: "Instagram media download",
+        aliases: ["instagram_media_download", "instagram reel download", "instagram post download", "ig media"],
+        credits: mcToCredits(MC_COSTS.instagram_media),
+        unit: "per post or reel",
+        notes: "Extracts post text, image URLs, and reel audio/video tracks, with local downloads when the server can write files."
+      },
+      {
+        key: "instagram_transcribe",
+        label: "Instagram media transcription",
+        aliases: ["instagram transcript", "instagram reel transcript", "ig transcribe"],
+        credits: mcToCredits(MC_COSTS.instagram_transcribe),
+        unit: "per call",
+        notes: "Whisper transcription of selected Instagram audio/video media via fal.ai."
+      },
+      {
+        key: "browser_minute",
+        label: "Interactive browser session",
+        aliases: ["browser_open", "browser agent", "browser_agent", "live browser", "browse", "browser control", "interactive browser"],
+        credits: mcToCredits(MC_COSTS.browser_minute),
+        unit: "per minute of use",
+        notes: "Metered per minute of use for the whole time the browser session is open. Close the session to stop the meter; abandoned sessions are auto-closed after a short idle window."
+      },
+      {
+        key: "reddit_thread",
+        label: "Reddit thread base lookup",
+        aliases: ["reddit", "reddit_thread", "reddit comments", "subreddit", "reddit post"],
+        credits: mcToCredits(MC_COSTS.reddit_thread),
+        unit: "per thread",
+        notes: "Base lookup for the post itself. Comments billed separately per comment at reddit_comment rate. Refunded if the thread cannot be retrieved."
+      },
+      {
+        key: "reddit_comment",
+        label: "Reddit comment",
+        aliases: ["reddit_comment", "reddit comments"],
+        credits: mcToCredits(MC_COSTS.reddit_comment),
+        unit: "per comment",
+        notes: "Charged per comment actually extracted, billed down automatically if the balance runs out mid-thread."
+      },
+      {
+        key: "video_analysis",
+        label: "Video breakdown (frame-by-frame + transcript)",
+        aliases: ["video_frame_analysis", "video breakdown", "video analysis", "analyze video"],
+        credits: mcToCredits(MC_COSTS.video_analysis),
+        unit: "per 120 frames (max 480)",
+        notes: "Full multi-lens video breakdown: samples frames across a video (up to 30 minutes), analyzes each with vision AI, transcribes the audio, then produces summary, pacing/energy, words-per-minute, topic outline, key points, hook analysis, visual style, and a how-to-replicate recipe with a quality-control pass. $1 per 120 frames requested (max 480 = $4); billed down automatically if the video cannot use the requested frames, and refunded fully if the run fails."
+      },
+      {
+        key: "trustpilot_reviews",
+        label: "Trustpilot review harvest",
+        aliases: ["trustpilot_reviews", "trustpilot", "trustpilot reviews"],
+        credits: mcToCredits(MC_COSTS.trustpilot_reviews),
+        unit: "per call",
+        notes: "Base lookup. Reviews billed separately per card at trustpilot_review rate. Refunded if no reviews are found."
+      },
+      {
+        key: "trustpilot_review",
+        label: "Trustpilot review card",
+        aliases: ["trustpilot_review", "trustpilot card"],
+        credits: mcToCredits(MC_COSTS.trustpilot_review),
+        unit: "per review card",
+        notes: "Charged per review actually extracted, billed down automatically if the balance runs out mid-page."
+      },
+      {
+        key: "g2_reviews",
+        label: "G2 review harvest",
+        aliases: ["g2_reviews", "g2", "g2 reviews"],
+        credits: mcToCredits(MC_COSTS.g2_reviews),
+        unit: "per call",
+        notes: "Base lookup. Reviews billed separately per card at g2_review rate. Refunded if no reviews are found."
+      },
+      {
+        key: "g2_review",
+        label: "G2 review card",
+        aliases: ["g2_review", "g2 card"],
+        credits: mcToCredits(MC_COSTS.g2_review),
+        unit: "per review card",
+        notes: "Charged per review actually extracted (each card carries up to 3 Q&A sections), billed down automatically if the balance runs out mid-page."
+      },
+      {
+        key: "diff_page",
+        label: "Page change check",
+        aliases: ["diff_page", "page diff", "change detection"],
+        credits: mcToCredits(MC_COSTS.diff_page),
+        unit: "per check",
+        notes: "Same cost as a single page extract \u2014 one scrape is performed per check, then compared against your last stored snapshot for that URL."
+      }
+    ];
+    CONCURRENCY_PRICE_ID = "price_1Ta1NRS8aAcsk3TGwsRnYbix";
+    SUBSCRIPTION_TIERS = {
+      "price_1TmiHRS8aAcsk3TGwmSNfNIa": { tier: "starter", label: "Starter", price_id: "price_1TmiHRS8aAcsk3TGwmSNfNIa", monthly_usd: 12, credits_mc: 8e6, concurrency: 3, intro_coupon: "mcp-starter-1dollar-intro-12" },
+      "price_1Tmg1nS8aAcsk3TGe8zcnGTM": { tier: "growth", label: "Growth", price_id: "price_1Tmg1nS8aAcsk3TGe8zcnGTM", monthly_usd: 100, credits_mc: 8e7, concurrency: 10, intro_coupon: null },
+      "price_1Tmg1nS8aAcsk3TGsZw34iXS": { tier: "scale", label: "Scale", price_id: "price_1Tmg1nS8aAcsk3TGsZw34iXS", monthly_usd: 250, credits_mc: 24e7, concurrency: 20, intro_coupon: null }
+    };
+    SUBSCRIPTION_TIER_BY_KEY = Object.fromEntries(
+      Object.values(SUBSCRIPTION_TIERS).map((t) => [t.tier, t])
+    );
+    CONCURRENCY_SLOT_PRICE_USD = 5;
+    CONCURRENCY_SLOT_PRICE_CURRENCY = "usd";
+    CONCURRENCY_SLOT_PRICE_INTERVAL = "month";
+    CONCURRENCY_SLOT_PRICE_LABEL = "$5/month";
+    CONCURRENCY_SLOT_TERMINAL_COMMAND = "npx -y -p mcp-scraper@latest mcp-scraper-cli billing concurrency checkout";
+    FREE_SIGNUP_MC = 1e5;
+    MEMORY_FREE_QUOTA_BYTES = 1e7 * 1;
+    MEMORY_PLANS = {
+      "price_1TnMSTS8aAcsk3TGBgiwuvqL": { plan: "pro", label: "Pro", price_id: "price_1TnMSTS8aAcsk3TGBgiwuvqL", interval: "month", monthly_usd: 19, quota_bytes: 5e9 },
+      "price_1TnMSTS8aAcsk3TGWBVU2agY": { plan: "pro", label: "Pro", price_id: "price_1TnMSTS8aAcsk3TGWBVU2agY", interval: "year", monthly_usd: 19, quota_bytes: 5e9 }
+    };
+    MEMORY_PLAN_QUOTA = { free: MEMORY_FREE_QUOTA_BYTES, pro: 5e9, team: 5e10 };
+    MEMORY_MARGIN_MULTIPLE = 3;
+    MEMORY_PRO_COST_BUDGET_USD = 19 / MEMORY_MARGIN_MULTIPLE;
+    SCHEDULING_PLANS = {
+      "price_1ToXcHS8aAcsk3TGf3cW7zHx": { price_id: "price_1ToXcHS8aAcsk3TGf3cW7zHx", label: "Scheduled Actions", interval: "month", monthly_usd: 10, quota_per_period: 1e3 }
+    };
+    LedgerOperation = {
+      TOPUP: "topup",
+      SUBSCRIPTION: "subscription",
+      SIGNUP_GRANT: "signup_grant",
+      MONTHLY_REFRESH: "monthly_free_refresh",
+      PAA: "paa",
+      PAA_REFUND: "paa_refund",
+      SERP: "serp",
+      REFUND: "refund",
+      TRANSCRIPTION: "transcription",
+      TRANSCRIPTION_HOLD: "transcription_hold",
+      TRANSCRIPTION_REFUND: "transcription_refund",
+      YT_CHANNEL: "yt_channel",
+      FB_AD: "fb_ad",
+      MAPS_SEARCH: "maps_search",
+      MAPS_PLACE: "maps_place",
+      MAPS_REVIEW: "maps_review",
+      MAPS_REVIEW_REFUND: "maps_review_refund",
+      EXTRACT_SITE: "extract_site",
+      EXTRACT_SITE_REFUND: "extract_site_refund",
+      EXTRACT_URL: "page_scrape",
+      URL_MAP: "url_map",
+      EXTRACT_SITE_HOLD: "extract_site_hold",
+      YT_CHANNEL_REFUND: "yt_channel_refund",
+      FB_AD_REFUND: "fb_ad_refund",
+      URL_MAP_REFUND: "url_map_refund",
+      FB_SEARCH: "fb_search",
+      FB_TRANSCRIBE: "fb_transcribe",
+      FB_SEARCH_REFUND: "fb_search_refund",
+      FB_TRANSCRIBE_REFUND: "fb_transcribe_refund",
+      GOOGLE_ADS_SEARCH: "google_ads_search",
+      GOOGLE_ADS_SEARCH_REFUND: "google_ads_search_refund",
+      GOOGLE_ADS_INTEL: "google_ads_intel",
+      GOOGLE_ADS_INTEL_REFUND: "google_ads_intel_refund",
+      GOOGLE_ADS_TRANSCRIBE: "google_ads_transcribe",
+      GOOGLE_ADS_TRANSCRIBE_REFUND: "google_ads_transcribe_refund",
+      INSTAGRAM_PROFILE: "instagram_profile",
+      INSTAGRAM_PROFILE_REFUND: "instagram_profile_refund",
+      INSTAGRAM_MEDIA: "instagram_media",
+      INSTAGRAM_MEDIA_REFUND: "instagram_media_refund",
+      INSTAGRAM_TRANSCRIBE: "instagram_transcribe",
+      INSTAGRAM_TRANSCRIBE_REFUND: "instagram_transcribe_refund",
+      BROWSER_SESSION: "browser_session",
+      REDDIT_THREAD: "reddit_thread",
+      REDDIT_THREAD_REFUND: "reddit_thread_refund",
+      REDDIT_COMMENT: "reddit_comment",
+      REDDIT_COMMENT_REFUND: "reddit_comment_refund",
+      VIDEO_ANALYSIS: "video_analysis",
+      VIDEO_ANALYSIS_REFUND: "video_analysis_refund",
+      VIDEO_ANALYSIS_ADJUST: "video_analysis_adjust",
+      VIDEO_ANALYSIS_LLM: "video_analysis_llm",
+      MEMORY_AI: "memory_ai",
+      MEMORY_AI_REFUND: "memory_ai_refund",
+      TRUSTPILOT_REVIEWS: "trustpilot_reviews",
+      TRUSTPILOT_REVIEWS_REFUND: "trustpilot_reviews_refund",
+      TRUSTPILOT_REVIEW: "trustpilot_review",
+      TRUSTPILOT_REVIEW_REFUND: "trustpilot_review_refund",
+      G2_REVIEWS: "g2_reviews",
+      G2_REVIEWS_REFUND: "g2_reviews_refund",
+      G2_REVIEW: "g2_review",
+      G2_REVIEW_REFUND: "g2_review_refund",
+      DIFF_PAGE: "diff_page",
+      DIFF_PAGE_REFUND: "diff_page_refund",
+      SITE_AUDIT_LLM: "site_audit_llm"
+    };
+    MEMORY_AI_MARGIN_MULTIPLE = 3;
+    SITE_AUDIT_LLM_MARGIN_MULTIPLE = 1.5;
+    VIDEO_ANALYSIS_LLM_MARGIN_MULTIPLE = 1.5;
+    MC_PER_USD = SUBSCRIPTION_TIERS["price_1TmiHRS8aAcsk3TGwmSNfNIa"].credits_mc / SUBSCRIPTION_TIERS["price_1TmiHRS8aAcsk3TGwmSNfNIa"].monthly_usd;
+  }
+});
+
 // src/services/site-architecture-auditor/prompts/ingest-validate.ts
 function buildIngestValidatePrompt(input) {
   const sessionPath = `<session-path-from-job>`;
@@ -8903,6 +9360,9 @@ var init_site_audit_service = __esm({
     import_p_limit = __toESM(require("p-limit"), 1);
     init_llm_parse_with_retry();
     init_schemas();
+    init_db();
+    init_rates();
+    init_cost_context();
     init_ingest_validate();
     init_enrich_competitor();
     init_build_graph_per_site();
@@ -8961,8 +9421,23 @@ var init_site_audit_service = __esm({
       async getJobsForUser(userId) {
         return this.deps.repo.listSiteAuditJobs(userId);
       }
+      async runLlmBilled(userId, source, fn) {
+        const before = this.deps.llm.totalCostUsd;
+        const result = await runWithCostContext({ ...currentCostContext(), op: "audit_site", userId }, fn);
+        const deltaCostUsd = this.deps.llm.totalCostUsd - before;
+        if (deltaCostUsd <= 0) return result;
+        const mc = Math.round(deltaCostUsd * SITE_AUDIT_LLM_MARGIN_MULTIPLE * MC_PER_USD);
+        if (mc <= 0) return result;
+        const { ok } = await debitMc(userId, mc, LedgerOperation.SITE_AUDIT_LLM, source);
+        if (!ok) {
+          console.warn(`[site-audit-service] SITE_AUDIT_LLM surcharge debit failed for user ${userId} (${source}): insufficient balance for ${mc}mc`);
+        }
+        return result;
+      }
       async runIngestPhase(jobId, payload) {
         const { request } = payload;
+        const job = await this.deps.repo.getSiteAuditJob(jobId);
+        if (!job) throw new Error(`[site-audit-service] runIngestPhase: job ${jobId} not found`);
         const input = {
           clientDomain: request.clientDomain,
           sfInternalPath: request.sfInternalPath,
@@ -8977,14 +9452,18 @@ var init_site_audit_service = __esm({
           businessContext: request.businessContext,
           priorAuditMemoryEnabled: request.priorAuditMemoryEnabled
         };
-        const result = await llmParseWithRetry(
-          ValidatedFileManifestSchema,
-          (attempt, error) => {
-            void attempt;
-            void error;
-            return buildIngestValidatePrompt(input);
-          },
-          this.deps.llm
+        const result = await this.runLlmBilled(
+          job.user_id,
+          `${jobId}:phase1-ingest`,
+          () => llmParseWithRetry(
+            ValidatedFileManifestSchema,
+            (attempt, error) => {
+              void attempt;
+              void error;
+              return buildIngestValidatePrompt(input);
+            },
+            this.deps.llm
+          )
         );
         await this.deps.repo.logSiteAuditPhaseComplete(jobId, "phase1-ingest", result);
       }
@@ -8999,23 +9478,27 @@ var init_site_audit_service = __esm({
         const competitorDomainsCsvPath = request.competitorDomainsCsvPath;
         void competitorDomainsCsvPath;
         const competitors = await this._readCompetitorDomains(request);
-        const results = await Promise.all(
-          competitors.map(
-            (domain, idx) => limit(async () => {
-              const competitorIndex = idx + 1;
-              const rawPrompt = buildEnrichCompetitorPrompt({ domain, sessionPath });
-              const prompt = rawPrompt.replace("<competitor-index>", String(competitorIndex));
-              const result = await llmParseWithRetry(
-                EnrichCompetitorOutputSchema,
-                (attempt, error) => {
-                  void attempt;
-                  void error;
-                  return prompt;
-                },
-                this.deps.llm
-              );
-              return result;
-            })
+        const results = await this.runLlmBilled(
+          job.user_id,
+          `${jobId}:phase1-enrich`,
+          () => Promise.all(
+            competitors.map(
+              (domain, idx) => limit(async () => {
+                const competitorIndex = idx + 1;
+                const rawPrompt = buildEnrichCompetitorPrompt({ domain, sessionPath });
+                const prompt = rawPrompt.replace("<competitor-index>", String(competitorIndex));
+                const result = await llmParseWithRetry(
+                  EnrichCompetitorOutputSchema,
+                  (attempt, error) => {
+                    void attempt;
+                    void error;
+                    return prompt;
+                  },
+                  this.deps.llm
+                );
+                return result;
+              })
+            )
           )
         );
         await this.deps.repo.logSiteAuditPhaseComplete(jobId, "phase1-ingest", { competitors: results });
@@ -9029,19 +9512,23 @@ var init_site_audit_service = __esm({
         const clientDomain = request.clientDomain;
         const urlInventory = [];
         const directedEdgeCount = 0;
-        const orphanAnnotation = await llmParseWithRetry(
-          OrphanAnnotationOutputSchema,
-          (attempt, error) => {
-            void attempt;
-            void error;
-            return buildGraphPerSitePrompt({
-              siteId: "client",
-              domain: clientDomain,
-              urlInventory,
-              directedEdgeCount
-            });
-          },
-          this.deps.llm
+        const orphanAnnotation = await this.runLlmBilled(
+          job.user_id,
+          `${jobId}:phase2-build-graph`,
+          () => llmParseWithRetry(
+            OrphanAnnotationOutputSchema,
+            (attempt, error) => {
+              void attempt;
+              void error;
+              return buildGraphPerSitePrompt({
+                siteId: "client",
+                domain: clientDomain,
+                urlInventory,
+                directedEdgeCount
+              });
+            },
+            this.deps.llm
+          )
         );
         const graphMetrics = await this.deps.python.runGraphMetrics({
           directedLinkGraphPath: `${sessionPath}/phase1/directed_link_graph.json`,
@@ -9086,50 +9573,52 @@ var init_site_audit_service = __esm({
         }
         if (currentBatch.length > 0) batches.push(currentBatch);
         const usZipsCsvPath = process.env["US_ZIPS_CSV_PATH"] ?? "/Users/vilovieta/Downloads/sales-magician-api-leads-magician-01c6cff78e31/tools/analytics/data/uszips.csv";
-        for (const batch of batches) {
-          let classified = [];
-          let unclassifiedUrls = batch.map((u) => u.url);
-          try {
-            const pythonResult = await this.deps.python.runLocationClassifier({
-              urls: batch.map((u) => u.url),
-              usZipsCsvPath
-            });
-            classified = pythonResult.classified;
-            unclassifiedUrls = pythonResult.unclassified;
-          } catch (err) {
-            console.warn(
-              "[site-audit-service] runClassifyPhase: python.runLocationClassifier failed (falling back to LLM):",
-              err instanceof Error ? err.message : String(err)
-            );
-            unclassifiedUrls = batch.map((u) => u.url);
-          }
-          for (const c of classified) {
-            allClassifications.push({
-              url: c.url,
-              page_type: c.pageType,
-              confidence: c.confidence > 0.6 ? "high" : c.confidence > 0.4 ? "medium" : "low",
-              _classification_label: "[COMPUTED:python/location-classifier]",
-              topic_cluster: null,
-              location_signals: c.locationSignals
-            });
-          }
-          if (unclassifiedUrls.length > 0) {
-            const unclassifiedBatch = batch.filter((u) => unclassifiedUrls.includes(u.url));
-            const llmResults = await llmParseWithRetry(
-              import_zod10.z.array(PageTypeClassificationRowSchema),
-              (attempt, error) => {
-                void attempt;
-                void error;
-                return buildMetricsClassifyPrompt({
-                  urls: unclassifiedBatch,
-                  siteDomain: clientDomain
-                });
-              },
-              this.deps.llm
-            );
-            allClassifications.push(...llmResults);
+        await this.runLlmBilled(job.user_id, `${jobId}:phase3-classify`, async () => {
+          for (const batch of batches) {
+            let classified = [];
+            let unclassifiedUrls = batch.map((u) => u.url);
+            try {
+              const pythonResult = await this.deps.python.runLocationClassifier({
+                urls: batch.map((u) => u.url),
+                usZipsCsvPath
+              });
+              classified = pythonResult.classified;
+              unclassifiedUrls = pythonResult.unclassified;
+            } catch (err) {
+              console.warn(
+                "[site-audit-service] runClassifyPhase: python.runLocationClassifier failed (falling back to LLM):",
+                err instanceof Error ? err.message : String(err)
+              );
+              unclassifiedUrls = batch.map((u) => u.url);
+            }
+            for (const c of classified) {
+              allClassifications.push({
+                url: c.url,
+                page_type: c.pageType,
+                confidence: c.confidence > 0.6 ? "high" : c.confidence > 0.4 ? "medium" : "low",
+                _classification_label: "[COMPUTED:python/location-classifier]",
+                topic_cluster: null,
+                location_signals: c.locationSignals
+              });
+            }
+            if (unclassifiedUrls.length > 0) {
+              const unclassifiedBatch = batch.filter((u) => unclassifiedUrls.includes(u.url));
+              const llmResults = await llmParseWithRetry(
+                import_zod10.z.array(PageTypeClassificationRowSchema),
+                (attempt, error) => {
+                  void attempt;
+                  void error;
+                  return buildMetricsClassifyPrompt({
+                    urls: unclassifiedBatch,
+                    siteDomain: clientDomain
+                  });
+                },
+                this.deps.llm
+              );
+              allClassifications.push(...llmResults);
+            }
           }
-        }
+        });
         void sessionPath;
         await this.deps.repo.logSiteAuditPhaseComplete(jobId, "phase3-classify", {
           classifications: allClassifications,
@@ -9142,20 +9631,24 @@ var init_site_audit_service = __esm({
         if (!job) throw new Error(`[site-audit-service] runComparePhase: job ${jobId} not found`);
         const request = JSON.parse(job.request);
         const clientDomain = request.clientDomain;
-        const result = await llmParseWithRetry(
-          CompareRecommendOutputSchema,
-          (attempt, error) => {
-            void attempt;
-            void error;
-            return buildCompareRecommendPrompt({
-              clientDomain,
-              clientMoneyPages: [],
-              competitorSites: [],
-              clientUrlMetrics: [],
-              clientPageTypes: []
-            });
-          },
-          this.deps.llm
+        const result = await this.runLlmBilled(
+          job.user_id,
+          `${jobId}:phase4-compare`,
+          () => llmParseWithRetry(
+            CompareRecommendOutputSchema,
+            (attempt, error) => {
+              void attempt;
+              void error;
+              return buildCompareRecommendPrompt({
+                clientDomain,
+                clientMoneyPages: [],
+                competitorSites: [],
+                clientUrlMetrics: [],
+                clientPageTypes: []
+              });
+            },
+            this.deps.llm
+          )
         );
         await this.deps.repo.logSiteAuditPhaseComplete(jobId, "phase4-compare", result);
         return result;
@@ -9185,20 +9678,24 @@ var init_site_audit_service = __esm({
           total: ilesTotal,
           components: ilesComponents
         };
-        const reportMarkdown = await llmParseWithRetry(
-          import_zod10.z.string(),
-          (attempt, error) => {
-            void attempt;
-            void error;
-            return buildScoreSynthesizePrompt({
-              preComputedArchitectureScore,
-              preComputedIlesScore,
-              rankedBacklog: [],
-              clientDomain,
-              sessionPath
-            });
-          },
-          this.deps.llm
+        const reportMarkdown = await this.runLlmBilled(
+          job.user_id,
+          `${jobId}:phase5-synthesize`,
+          () => llmParseWithRetry(
+            import_zod10.z.string(),
+            (attempt, error) => {
+              void attempt;
+              void error;
+              return buildScoreSynthesizePrompt({
+                preComputedArchitectureScore,
+                preComputedIlesScore,
+                rankedBacklog: [],
+                clientDomain,
+                sessionPath
+              });
+            },
+            this.deps.llm
+          )
         );
         const auditResult = SiteAuditResultSchema.parse({
           jobId,
@@ -9983,364 +10480,6 @@ var init_image_audit = __esm({
   }
 });
 
-// src/api/rates.ts
-function browserActiveCostMc(activeMs) {
-  return Math.round(activeMs * MC_PER_BROWSER_MS);
-}
-function concurrencySlotBillingInfo() {
-  const billingUrl = process.env.CONCURRENCY_BILLING_URL ?? "https://mcpscraper.dev/billing";
-  return {
-    product: "Extra concurrency slot",
-    price_label: CONCURRENCY_SLOT_PRICE_LABEL,
-    unit_amount_usd: CONCURRENCY_SLOT_PRICE_USD,
-    currency: CONCURRENCY_SLOT_PRICE_CURRENCY,
-    interval: CONCURRENCY_SLOT_PRICE_INTERVAL,
-    billing_url: billingUrl,
-    terminal_command: CONCURRENCY_SLOT_TERMINAL_COMMAND,
-    terminal_command_with_api_key_env: `MCP_SCRAPER_API_KEY=sk_live_your_key ${CONCURRENCY_SLOT_TERMINAL_COMMAND}`
-  };
-}
-function mcToCredits(mc) {
-  return mc / MC_PER_CREDIT;
-}
-function insufficientBalanceResponse(balanceMc, requiredMc) {
-  const topupUrl = process.env.TOPUP_URL ?? "https://mcpscraper.dev/billing";
-  const balanceCredits = mcToCredits(balanceMc);
-  const requiredCredits = mcToCredits(requiredMc);
-  return {
-    error: "insufficient_balance",
-    error_code: "insufficient_balance",
-    message: `Insufficient credits. Balance: ${balanceCredits} credits. This call requires ${requiredCredits} credits. Top up at ${topupUrl}`,
-    balance_credits: balanceCredits,
-    required_credits: requiredCredits,
-    topup_url: topupUrl
-  };
-}
-var MC_COSTS, MC_PER_BROWSER_MS, BROWSER_OPEN_MIN_BALANCE_MC, MC_PER_CREDIT, CREDIT_COST_CATALOG, CONCURRENCY_PRICE_ID, SUBSCRIPTION_TIERS, SUBSCRIPTION_TIER_BY_KEY, CONCURRENCY_SLOT_PRICE_USD, CONCURRENCY_SLOT_PRICE_CURRENCY, CONCURRENCY_SLOT_PRICE_INTERVAL, CONCURRENCY_SLOT_PRICE_LABEL, CONCURRENCY_SLOT_TERMINAL_COMMAND, FREE_SIGNUP_MC, MEMORY_FREE_QUOTA_BYTES, MEMORY_PLANS, MEMORY_PLAN_QUOTA, MEMORY_MARGIN_MULTIPLE, MEMORY_PRO_COST_BUDGET_USD, SCHEDULING_PLANS, LedgerOperation, MEMORY_AI_MARGIN_MULTIPLE, MC_PER_USD;
-var init_rates = __esm({
-  "src/api/rates.ts"() {
-    "use strict";
-    MC_COSTS = {
-      serp: 100,
-      paa: 200,
-      paa_base: 1e3,
-      page_scrape: 100,
-      url_map: 500,
-      yt_channel: 200,
-      yt_transcription: 200,
-      fb_ad: 200,
-      maps_search: 500,
-      maps_place: 3e3,
-      maps_review: 100,
-      fb_search: 600,
-      fb_transcribe: 200,
-      google_ads_search: 600,
-      google_ads_intel: 200,
-      google_ads_transcribe: 200,
-      instagram_profile: 400,
-      instagram_media: 400,
-      instagram_transcribe: 50,
-      browser_minute: 12e3,
-      reddit_thread: 2e3,
-      video_analysis: 666667,
-      trustpilot_reviews: 500,
-      trustpilot_review: 100,
-      g2_reviews: 500,
-      g2_review: 150
-    };
-    MC_PER_BROWSER_MS = MC_COSTS.browser_minute / 6e4;
-    BROWSER_OPEN_MIN_BALANCE_MC = 1e3;
-    MC_PER_CREDIT = 1e3;
-    CREDIT_COST_CATALOG = [
-      {
-        key: "serp",
-        label: "SERP search",
-        aliases: ["search_serp", "serp", "google search", "organic results"],
-        credits: mcToCredits(MC_COSTS.serp),
-        unit: "per search",
-        notes: "Returns AI Overview, PAA snippet, videos, forums, and local pack."
-      },
-      {
-        key: "paa",
-        label: "PAA harvest",
-        aliases: ["harvest_paa", "paa", "people also ask", "questions"],
-        credits: mcToCredits(MC_COSTS.paa),
-        unit: `per question (+${mcToCredits(MC_COSTS.paa_base)} credit base)`,
-        notes: `Full SERP feature extraction. Billed ${mcToCredits(MC_COSTS.paa_base)} credit base + ${mcToCredits(MC_COSTS.paa)} per question actually returned (unused estimate refunded).`
-      },
-      {
-        key: "page_scrape",
-        label: "Page crawl / extract",
-        aliases: ["extract_url", "extract_site", "page scrape", "url scrape", "single page", "site crawl"],
-        credits: mcToCredits(MC_COSTS.page_scrape),
-        unit: "per page",
-        notes: "Applies to both single-URL extraction and per-page site crawls."
-      },
-      {
-        key: "url_map",
-        label: "Site URL mapping",
-        aliases: ["map_site_urls", "url map", "site map", "crawl urls"],
-        credits: mcToCredits(MC_COSTS.url_map),
-        unit: "per mapping operation",
-        notes: "Flat rate for the full /map-urls call regardless of URL count discovered."
-      },
-      {
-        key: "yt_channel",
-        label: "YouTube search / channel harvest",
-        aliases: ["youtube_harvest", "youtube search", "youtube channel", "yt_channel"],
-        credits: mcToCredits(MC_COSTS.yt_channel),
-        unit: "per call"
-      },
-      {
-        key: "yt_transcription",
-        label: "YouTube transcription",
-        aliases: ["youtube_transcribe", "youtube transcript", "transcription", "yt_transcription"],
-        credits: mcToCredits(MC_COSTS.yt_transcription),
-        unit: "per minute",
-        notes: "A 5-minute hold is taken, then reconciled to actual video duration."
-      },
-      {
-        key: "fb_ad",
-        label: "Facebook search / ad lookup",
-        aliases: ["facebook_page_intel", "facebook_ad_search", "facebook_ad", "facebook ads", "fb ads"],
-        credits: mcToCredits(MC_COSTS.fb_ad),
-        unit: "per call"
-      },
-      {
-        key: "maps_search",
-        label: "Maps business search",
-        aliases: ["maps_search", "google maps search", "gmb search", "gbp search", "business profiles"],
-        credits: mcToCredits(MC_COSTS.maps_search),
-        unit: "per search",
-        notes: "Returns up to 50 Google Maps business/profile candidates. Use maps_place_intel to hydrate selected businesses."
-      },
-      {
-        key: "maps_place",
-        label: "Maps business lookup",
-        aliases: ["maps_place_intel", "google maps", "maps place", "place intel"],
-        credits: mcToCredits(MC_COSTS.maps_place),
-        unit: "per business",
-        notes: "Base lookup. Reviews billed separately per card at maps_review rate."
-      },
-      {
-        key: "maps_review",
-        label: "Maps review",
-        aliases: ["maps_reviews", "google reviews", "review cards", "reviews"],
-        credits: mcToCredits(MC_COSTS.maps_review),
-        unit: "per review card",
-        notes: "Charged after extraction when includeReviews is true."
-      },
-      {
-        key: "fb_search",
-        label: "Facebook ad library search",
-        aliases: ["facebook_search", "fb_search", "fb ad search"],
-        credits: mcToCredits(MC_COSTS.fb_search),
-        unit: "per search",
-        notes: "Browser automation to search Facebook Ads Library by keyword."
-      },
-      {
-        key: "google_ads_search",
-        label: "Google Ads Transparency search",
-        aliases: ["google_ads_search", "google ads search", "ads transparency search"],
-        credits: mcToCredits(MC_COSTS.google_ads_search),
-        unit: "per search",
-        notes: "Browser automation to find advertisers in Google Ads Transparency Center by domain or name."
-      },
-      {
-        key: "google_ads_intel",
-        label: "Google Ads Transparency advertiser intel",
-        aliases: ["google_ads_page_intel", "google ads intel", "ads transparency intel"],
-        credits: mcToCredits(MC_COSTS.google_ads_intel),
-        unit: "per call",
-        notes: "Lists and hydrates an advertiser's creatives with image URLs and video references."
-      },
-      {
-        key: "google_ads_transcribe",
-        label: "Google ad video transcription",
-        aliases: ["google_ads_transcribe", "google ad transcript"],
-        credits: mcToCredits(MC_COSTS.google_ads_transcribe),
-        unit: "per minute",
-        notes: "A hold is taken, then reconciled to actual video duration."
-      },
-      {
-        key: "fb_transcribe",
-        label: "Facebook video / ad transcription",
-        aliases: ["facebook_transcribe", "facebook_video_transcribe", "fb_transcribe", "fb ad transcript"],
-        credits: mcToCredits(MC_COSTS.fb_transcribe),
-        unit: "per minute",
-        notes: "A hold is taken, then reconciled to actual video duration."
-      },
-      {
-        key: "instagram_profile",
-        label: "Instagram profile content discovery",
-        aliases: ["instagram_profile_content", "instagram profile", "instagram content list", "ig profile"],
-        credits: mcToCredits(MC_COSTS.instagram_profile),
-        unit: "per profile scan",
-        notes: "Browser extraction of public Instagram profile grid links. Complete history may require a logged-in profile."
-      },
-      {
-        key: "instagram_media",
-        label: "Instagram media download",
-        aliases: ["instagram_media_download", "instagram reel download", "instagram post download", "ig media"],
-        credits: mcToCredits(MC_COSTS.instagram_media),
-        unit: "per post or reel",
-        notes: "Extracts post text, image URLs, and reel audio/video tracks, with local downloads when the server can write files."
-      },
-      {
-        key: "instagram_transcribe",
-        label: "Instagram media transcription",
-        aliases: ["instagram transcript", "instagram reel transcript", "ig transcribe"],
-        credits: mcToCredits(MC_COSTS.instagram_transcribe),
-        unit: "per call",
-        notes: "Whisper transcription of selected Instagram audio/video media via fal.ai."
-      },
-      {
-        key: "browser_minute",
-        label: "Interactive browser session",
-        aliases: ["browser_open", "browser agent", "browser_agent", "live browser", "browse", "browser control", "interactive browser"],
-        credits: mcToCredits(MC_COSTS.browser_minute),
-        unit: "per minute of use",
-        notes: "Metered per minute of use for the whole time the browser session is open. Close the session to stop the meter; abandoned sessions are auto-closed after a short idle window."
-      },
-      {
-        key: "reddit_thread",
-        label: "Reddit thread + comments",
-        aliases: ["reddit", "reddit_thread", "reddit comments", "subreddit", "reddit post"],
-        credits: mcToCredits(MC_COSTS.reddit_thread),
-        unit: "per thread",
-        notes: "Captures a Reddit post and its comment tree. Charged on success; refunded if the thread cannot be retrieved."
-      },
-      {
-        key: "video_analysis",
-        label: "Video breakdown (frame-by-frame + transcript)",
-        aliases: ["video_frame_analysis", "video breakdown", "video analysis", "analyze video"],
-        credits: mcToCredits(MC_COSTS.video_analysis),
-        unit: "per 120 frames (max 480)",
-        notes: "Full multi-lens video breakdown: samples frames across a video (up to 30 minutes), analyzes each with vision AI, transcribes the audio, then produces summary, pacing/energy, words-per-minute, topic outline, key points, hook analysis, visual style, and a how-to-replicate recipe with a quality-control pass. $1 per 120 frames requested (max 480 = $4); billed down automatically if the video cannot use the requested frames, and refunded fully if the run fails."
-      },
-      {
-        key: "trustpilot_reviews",
-        label: "Trustpilot review harvest",
-        aliases: ["trustpilot_reviews", "trustpilot", "trustpilot reviews"],
-        credits: mcToCredits(MC_COSTS.trustpilot_reviews),
-        unit: "per call",
-        notes: "Base lookup. Reviews billed separately per card at trustpilot_review rate. Refunded if no reviews are found."
-      },
-      {
-        key: "trustpilot_review",
-        label: "Trustpilot review card",
-        aliases: ["trustpilot_review", "trustpilot card"],
-        credits: mcToCredits(MC_COSTS.trustpilot_review),
-        unit: "per review card",
-        notes: "Charged per review actually extracted, billed down automatically if the balance runs out mid-page."
-      },
-      {
-        key: "g2_reviews",
-        label: "G2 review harvest",
-        aliases: ["g2_reviews", "g2", "g2 reviews"],
-        credits: mcToCredits(MC_COSTS.g2_reviews),
-        unit: "per call",
-        notes: "Base lookup. Reviews billed separately per card at g2_review rate. Refunded if no reviews are found."
-      },
-      {
-        key: "g2_review",
-        label: "G2 review card",
-        aliases: ["g2_review", "g2 card"],
-        credits: mcToCredits(MC_COSTS.g2_review),
-        unit: "per review card",
-        notes: "Charged per review actually extracted (each card carries up to 3 Q&A sections), billed down automatically if the balance runs out mid-page."
-      }
-    ];
-    CONCURRENCY_PRICE_ID = "price_1Ta1NRS8aAcsk3TGwsRnYbix";
-    SUBSCRIPTION_TIERS = {
-      "price_1TmiHRS8aAcsk3TGwmSNfNIa": { tier: "starter", label: "Starter", price_id: "price_1TmiHRS8aAcsk3TGwmSNfNIa", monthly_usd: 12, credits_mc: 8e6, concurrency: 3, intro_coupon: "mcp-starter-1dollar-intro-12" },
-      "price_1Tmg1nS8aAcsk3TGe8zcnGTM": { tier: "growth", label: "Growth", price_id: "price_1Tmg1nS8aAcsk3TGe8zcnGTM", monthly_usd: 100, credits_mc: 8e7, concurrency: 10, intro_coupon: null },
-      "price_1Tmg1nS8aAcsk3TGsZw34iXS": { tier: "scale", label: "Scale", price_id: "price_1Tmg1nS8aAcsk3TGsZw34iXS", monthly_usd: 250, credits_mc: 24e7, concurrency: 20, intro_coupon: null }
-    };
-    SUBSCRIPTION_TIER_BY_KEY = Object.fromEntries(
-      Object.values(SUBSCRIPTION_TIERS).map((t) => [t.tier, t])
-    );
-    CONCURRENCY_SLOT_PRICE_USD = 5;
-    CONCURRENCY_SLOT_PRICE_CURRENCY = "usd";
-    CONCURRENCY_SLOT_PRICE_INTERVAL = "month";
-    CONCURRENCY_SLOT_PRICE_LABEL = "$5/month";
-    CONCURRENCY_SLOT_TERMINAL_COMMAND = "npx -y -p mcp-scraper@latest mcp-scraper-cli billing concurrency checkout";
-    FREE_SIGNUP_MC = 1e5;
-    MEMORY_FREE_QUOTA_BYTES = 1e7 * 1;
-    MEMORY_PLANS = {
-      "price_1TnMSTS8aAcsk3TGBgiwuvqL": { plan: "pro", label: "Pro", price_id: "price_1TnMSTS8aAcsk3TGBgiwuvqL", interval: "month", monthly_usd: 19, quota_bytes: 5e9 },
-      "price_1TnMSTS8aAcsk3TGWBVU2agY": { plan: "pro", label: "Pro", price_id: "price_1TnMSTS8aAcsk3TGWBVU2agY", interval: "year", monthly_usd: 19, quota_bytes: 5e9 }
-    };
-    MEMORY_PLAN_QUOTA = { free: MEMORY_FREE_QUOTA_BYTES, pro: 5e9, team: 5e10 };
-    MEMORY_MARGIN_MULTIPLE = 3;
-    MEMORY_PRO_COST_BUDGET_USD = 19 / MEMORY_MARGIN_MULTIPLE;
-    SCHEDULING_PLANS = {
-      "price_1ToXcHS8aAcsk3TGf3cW7zHx": { price_id: "price_1ToXcHS8aAcsk3TGf3cW7zHx", label: "Scheduled Actions", interval: "month", monthly_usd: 10, quota_per_period: 1e3 }
-    };
-    LedgerOperation = {
-      TOPUP: "topup",
-      SUBSCRIPTION: "subscription",
-      SIGNUP_GRANT: "signup_grant",
-      MONTHLY_REFRESH: "monthly_free_refresh",
-      PAA: "paa",
-      PAA_REFUND: "paa_refund",
-      SERP: "serp",
-      REFUND: "refund",
-      TRANSCRIPTION: "transcription",
-      TRANSCRIPTION_HOLD: "transcription_hold",
-      TRANSCRIPTION_REFUND: "transcription_refund",
-      YT_CHANNEL: "yt_channel",
-      FB_AD: "fb_ad",
-      MAPS_SEARCH: "maps_search",
-      MAPS_PLACE: "maps_place",
-      MAPS_REVIEW: "maps_review",
-      MAPS_REVIEW_REFUND: "maps_review_refund",
-      EXTRACT_SITE: "extract_site",
-      EXTRACT_SITE_REFUND: "extract_site_refund",
-      EXTRACT_URL: "page_scrape",
-      URL_MAP: "url_map",
-      EXTRACT_SITE_HOLD: "extract_site_hold",
-      YT_CHANNEL_REFUND: "yt_channel_refund",
-      FB_AD_REFUND: "fb_ad_refund",
-      URL_MAP_REFUND: "url_map_refund",
-      FB_SEARCH: "fb_search",
-      FB_TRANSCRIBE: "fb_transcribe",
-      FB_SEARCH_REFUND: "fb_search_refund",
-      FB_TRANSCRIBE_REFUND: "fb_transcribe_refund",
-      GOOGLE_ADS_SEARCH: "google_ads_search",
-      GOOGLE_ADS_SEARCH_REFUND: "google_ads_search_refund",
-      GOOGLE_ADS_INTEL: "google_ads_intel",
-      GOOGLE_ADS_INTEL_REFUND: "google_ads_intel_refund",
-      GOOGLE_ADS_TRANSCRIBE: "google_ads_transcribe",
-      GOOGLE_ADS_TRANSCRIBE_REFUND: "google_ads_transcribe_refund",
-      INSTAGRAM_PROFILE: "instagram_profile",
-      INSTAGRAM_PROFILE_REFUND: "instagram_profile_refund",
-      INSTAGRAM_MEDIA: "instagram_media",
-      INSTAGRAM_MEDIA_REFUND: "instagram_media_refund",
-      INSTAGRAM_TRANSCRIBE: "instagram_transcribe",
-      INSTAGRAM_TRANSCRIBE_REFUND: "instagram_transcribe_refund",
-      BROWSER_SESSION: "browser_session",
-      REDDIT_THREAD: "reddit_thread",
-      REDDIT_THREAD_REFUND: "reddit_thread_refund",
-      VIDEO_ANALYSIS: "video_analysis",
-      VIDEO_ANALYSIS_REFUND: "video_analysis_refund",
-      VIDEO_ANALYSIS_ADJUST: "video_analysis_adjust",
-      MEMORY_AI: "memory_ai",
-      MEMORY_AI_REFUND: "memory_ai_refund",
-      TRUSTPILOT_REVIEWS: "trustpilot_reviews",
-      TRUSTPILOT_REVIEWS_REFUND: "trustpilot_reviews_refund",
-      TRUSTPILOT_REVIEW: "trustpilot_review",
-      TRUSTPILOT_REVIEW_REFUND: "trustpilot_review_refund",
-      G2_REVIEWS: "g2_reviews",
-      G2_REVIEWS_REFUND: "g2_reviews_refund",
-      G2_REVIEW: "g2_review",
-      G2_REVIEW_REFUND: "g2_review_refund"
-    };
-    MEMORY_AI_MARGIN_MULTIPLE = 3;
-    MC_PER_USD = SUBSCRIPTION_TIERS["price_1TmiHRS8aAcsk3TGwmSNfNIa"].credits_mc / SUBSCRIPTION_TIERS["price_1TmiHRS8aAcsk3TGwmSNfNIa"].monthly_usd;
-  }
-});
-
 // src/api/site-extract-repository.ts
 var site_extract_repository_exports = {};
 __export(site_extract_repository_exports, {
@@ -11560,6 +11699,8 @@ var init_internal_memory_routes = __esm({
     import_hono2 = require("hono");
     init_db();
     init_rates();
+    init_cost_telemetry();
+    init_cost_context();
     internalMemoryApp = new import_hono2.Hono();
     internalMemoryApp.post("/ai-debit", async (c) => {
       const secret2 = c.req.header("x-internal-secret");
@@ -11572,6 +11713,11 @@ var init_internal_memory_routes = __esm({
       }
       const user = await getUserByEmail(body.identity.trim().toLowerCase());
       if (!user) return c.json({ ok: false, reason: "no_user" }, 200);
+      if (body.costUsd > 0) {
+        await runWithCostContext({ ...currentCostContext(), op: "memory_ai", userId: user.id }, async () => {
+          void recordVendorUsage({ vendor: "mcp_memory_ai", model: null, units: body.costUsd, unitType: "usd", error: null });
+        });
+      }
       const mc = Math.round(body.costUsd * MEMORY_AI_MARGIN_MULTIPLE * MC_PER_USD);
       if (mc <= 0) return c.json({ ok: true, mc: 0, balance_mc: null }, 200);
       const { ok, balance_mc } = await debitMc(user.id, mc, LedgerOperation.MEMORY_AI, `${body.source ?? "ai"}:${body.op ?? ""}`);
@@ -12834,6 +12980,563 @@ var init_youtube_harvest = __esm({
   }
 });
 
+// src/locations.ts
+var LOCATIONS;
+var init_locations = __esm({
+  "src/locations.ts"() {
+    "use strict";
+    LOCATIONS = {
+      "austin": "Austin,Texas,United States",
+      "new york": "New York,New York,United States",
+      "new york city": "New York,New York,United States",
+      "nyc": "New York,New York,United States",
+      "los angeles": "Los Angeles,California,United States",
+      "la": "Los Angeles,California,United States",
+      "chicago": "Chicago,Illinois,United States",
+      "houston": "Houston,Texas,United States",
+      "phoenix": "Phoenix,Arizona,United States",
+      "philadelphia": "Philadelphia,Pennsylvania,United States",
+      "philly": "Philadelphia,Pennsylvania,United States",
+      "san antonio": "San Antonio,Texas,United States",
+      "dallas": "Dallas,Texas,United States",
+      "miami": "Miami,Florida,United States",
+      "seattle": "Seattle,Washington,United States",
+      "denver": "Denver,Colorado,United States",
+      "loveland": "Loveland,Colorado,United States",
+      "loveland co": "Loveland,Colorado,United States",
+      "fort collins": "Fort Collins,Colorado,United States",
+      "boulder": "Boulder,Colorado,United States",
+      "colorado springs": "Colorado Springs,Colorado,United States",
+      "boston": "Boston,Massachusetts,United States",
+      "atlanta": "Atlanta,Georgia,United States",
+      "san francisco": "San Francisco,California,United States",
+      "sf": "San Francisco,California,United States",
+      "portland": "Portland,Oregon,United States",
+      "las vegas": "Las Vegas,Nevada,United States",
+      "minneapolis": "Minneapolis,Minnesota,United States",
+      "detroit": "Detroit,Michigan,United States",
+      "nashville": "Nashville,Tennessee,United States",
+      "charlotte": "Charlotte,North Carolina,United States",
+      "orlando": "Orlando,Florida,United States",
+      "san diego": "San Diego,California,United States",
+      "baltimore": "Baltimore,Maryland,United States",
+      "sacramento": "Sacramento,California,United States",
+      "columbus": "Columbus,Ohio,United States",
+      "indianapolis": "Indianapolis,Indiana,United States",
+      "san jose": "San Jose,California,United States",
+      "fort worth": "Fort Worth,Texas,United States",
+      "jacksonville": "Jacksonville,Florida,United States",
+      "memphis": "Memphis,Tennessee,United States",
+      "louisville": "Louisville,Kentucky,United States",
+      "raleigh": "Raleigh,North Carolina,United States",
+      "richmond": "Richmond,Virginia,United States",
+      "salt lake city": "Salt Lake City,Utah,United States",
+      "toronto": "Toronto,Ontario,Canada",
+      "vancouver": "Vancouver,British Columbia,Canada",
+      "montreal": "Montreal,Quebec,Canada",
+      "calgary": "Calgary,Alberta,Canada",
+      "ottawa": "Ottawa,Ontario,Canada",
+      "london": "London,England,United Kingdom",
+      "manchester": "Manchester,England,United Kingdom",
+      "birmingham": "Birmingham,England,United Kingdom",
+      "edinburgh": "Edinburgh,Scotland,United Kingdom",
+      "glasgow": "Glasgow,Scotland,United Kingdom",
+      "leeds": "Leeds,England,United Kingdom",
+      "sydney": "Sydney,New South Wales,Australia",
+      "melbourne": "Melbourne,Victoria,Australia",
+      "brisbane": "Brisbane,Queensland,Australia",
+      "perth": "Perth,Western Australia,Australia",
+      "adelaide": "Adelaide,South Australia,Australia",
+      "dublin": "Dublin,Leinster,Ireland"
+    };
+  }
+});
+
+// src/uule.ts
+function encodeVarint(value) {
+  const bytes = [];
+  let remaining = value;
+  do {
+    let byte = remaining & 127;
+    remaining >>>= 7;
+    if (remaining > 0) byte |= 128;
+    bytes.push(byte);
+  } while (remaining > 0);
+  return bytes;
+}
+function encodeUule(name) {
+  const locationBytes = Buffer.from(name, "utf8");
+  const payload = Buffer.concat([
+    Buffer.from([8, 2, 16, 32, 34]),
+    Buffer.from(encodeVarint(locationBytes.length)),
+    locationBytes
+  ]);
+  return `w+${payload.toString("base64")}`;
+}
+function normalizeLocation(input) {
+  const raw = input.toLowerCase().trim();
+  if (LOCATIONS[raw]) return LOCATIONS[raw];
+  const beforeComma = raw.split(",")[0].trim();
+  if (beforeComma !== raw && LOCATIONS[beforeComma]) return LOCATIONS[beforeComma];
+  const withoutState = raw.replace(/\s+[a-z]{2}$/, "").trim();
+  if (withoutState !== raw && LOCATIONS[withoutState]) return LOCATIONS[withoutState];
+  return input;
+}
+var init_uule = __esm({
+  "src/uule.ts"() {
+    "use strict";
+    init_locations();
+  }
+});
+
+// src/kernel-proxy-resolver.ts
+function proxyIdSuffix2(proxyId) {
+  return proxyId ? proxyId.slice(-6) : null;
+}
+function resolution(source, proxyMode, proxyId, target, error, disposable = false) {
+  const proxyType = source === "disabled" ? "none" : source === "configured_fallback" ? "configured" : source === "isp_created" ? "isp" : source === "location_created" || source === "location_reused" ? "residential" : "unknown";
+  return {
+    kernelProxyId: proxyId,
+    ...disposable && proxyId ? { disposableProxyId: proxyId } : {},
+    resolution: {
+      source,
+      proxyType,
+      proxyMode,
+      proxyIdPresent: Boolean(proxyId),
+      proxyIdSuffix: proxyIdSuffix2(proxyId),
+      target,
+      error,
+      disposable
+    }
+  };
+}
+function normalizeStateName(value) {
+  return value.trim().toLowerCase().replace(/\s+/g, " ");
+}
+function normalizeCountryName(value) {
+  return value.trim().toLowerCase().replace(/\./g, "").replace(/\s+/g, " ");
+}
+function isUnitedStates(country) {
+  if (!country) return true;
+  const normalized = normalizeCountryName(country);
+  return normalized === "united states" || normalized === "united states of america" || normalized === "usa" || normalized === "us";
+}
+function stateCodeFor(region) {
+  const trimmed = region.trim();
+  if (/^[A-Za-z]{2}$/.test(trimmed)) return trimmed.toUpperCase();
+  return US_STATE_CODES[normalizeStateName(trimmed)] ?? null;
+}
+function kernelCityIdentifierCandidates(city) {
+  const ascii = city.normalize("NFKD").replace(/[^\x00-\x7F]/g, "").toLowerCase();
+  const words = ascii.split(/[^a-z0-9]+/).filter(Boolean);
+  const underscored = words.join("_");
+  const compact = words.join("");
+  return Array.from(new Set([underscored, compact].filter(Boolean)));
+}
+function proxyName(country, state, city) {
+  return city ? `mcp-serp-residential-${country.toLowerCase()}-${state.toLowerCase()}-${city}` : `mcp-serp-residential-${country.toLowerCase()}-${state.toLowerCase()}`;
+}
+function freshProxyName(baseName, attemptIndex) {
+  const stamp = `${Date.now()}-${attemptIndex ?? 0}-${Math.random().toString(36).slice(2, 8)}`;
+  return `${baseName}-fresh-${stamp}`;
+}
+function zipProxyName(zip) {
+  return `mcp-serp-residential-us-zip-${zip}`;
+}
+function parseKernelLocationProxyTarget(location2, gl) {
+  if (!location2 || gl.toLowerCase() !== "us") return null;
+  const canonicalLocation = normalizeLocation(location2);
+  let parts = canonicalLocation.split(",").map((part) => part.trim()).filter(Boolean);
+  if (parts.length > 1 && isUnitedStates(parts[parts.length - 1])) {
+    parts = parts.slice(0, -1);
+  }
+  if (parts.length === 1) {
+    const stateOnly = stateCodeFor(parts[0]);
+    if (!stateOnly) return null;
+    return {
+      canonicalLocation,
+      level: "state",
+      country: "US",
+      state: stateOnly,
+      city: "",
+      cityCandidates: [],
+      proxyName: proxyName("US", stateOnly),
+      config: {
+        country: "US",
+        state: stateOnly
+      }
+    };
+  }
+  const [city = "", region = ""] = parts;
+  if (!city || !region) return null;
+  const state = stateCodeFor(region);
+  if (!state) return null;
+  const cityCandidates = kernelCityIdentifierCandidates(city);
+  const primaryCity = cityCandidates[0];
+  if (!primaryCity) return null;
+  return {
+    canonicalLocation,
+    level: "city",
+    country: "US",
+    state,
+    city: primaryCity,
+    cityCandidates,
+    proxyName: proxyName("US", state, primaryCity),
+    config: {
+      country: "US",
+      state,
+      city: primaryCity
+    }
+  };
+}
+function cityZipKey(target) {
+  return `${target.city}|${target.state}`;
+}
+function knownZipFor(target, explicitZip) {
+  if (explicitZip && /^\d{5}$/.test(explicitZip)) return explicitZip;
+  return US_CITY_CENTER_ZIPS[cityZipKey(target)] ?? null;
+}
+function zipTarget(target, zip) {
+  return {
+    ...target,
+    level: "zip",
+    zip,
+    proxyName: zipProxyName(zip),
+    config: {
+      country: target.country,
+      state: target.state,
+      zip
+    }
+  };
+}
+function withProxyName(target, name) {
+  return {
+    ...target,
+    proxyName: name
+  };
+}
+function configMatches(config, target, city) {
+  if (target.level === "zip") {
+    return config?.country?.toUpperCase() === target.country && config?.zip === target.zip;
+  }
+  return config?.country?.toUpperCase() === target.country && config?.state?.toUpperCase() === target.state && (city ? config?.city === city : !config?.city);
+}
+function findExistingTargetProxy(proxies, target) {
+  return proxies.find((proxy) => proxy.type === "residential" && proxy.status !== "unavailable" && Boolean(proxy.id) && (proxy.name === target.proxyName || configMatches(proxy.config, target, target.level === "city" ? target.city : void 0))) ?? null;
+}
+function findExistingProxy(proxies, target) {
+  for (const city of target.cityCandidates) {
+    const name = proxyName(target.country, target.state, city);
+    const found = proxies.find((proxy) => proxy.type === "residential" && proxy.status !== "unavailable" && Boolean(proxy.id) && (proxy.name === name || configMatches(proxy.config, target, city)));
+    if (found) return found;
+  }
+  return null;
+}
+function stateTarget(target) {
+  return {
+    ...target,
+    level: "state",
+    proxyName: proxyName(target.country, target.state),
+    config: {
+      country: target.country,
+      state: target.state
+    }
+  };
+}
+function findExistingStateProxy(proxies, target) {
+  const name = proxyName(target.country, target.state);
+  return proxies.find((proxy) => proxy.type === "residential" && proxy.status !== "unavailable" && Boolean(proxy.id) && (proxy.name === name || configMatches(proxy.config, target))) ?? null;
+}
+function escalatedTargetLevel(target, attemptIndex) {
+  return stateTarget(target);
+}
+function errorText2(err) {
+  return err instanceof Error ? err.message : String(err);
+}
+function freshTargetCandidates(target, explicitZip, attemptIndex) {
+  const out = [];
+  const zip = knownZipFor(target, explicitZip);
+  if (zip) {
+    const targetZip = zipTarget(target, zip);
+    out.push(withProxyName(targetZip, freshProxyName(targetZip.proxyName, attemptIndex)));
+  }
+  for (const city of target.cityCandidates) {
+    const cityTarget = {
+      ...target,
+      level: "city",
+      city,
+      proxyName: proxyName(target.country, target.state, city),
+      config: {
+        country: target.country,
+        state: target.state,
+        city
+      }
+    };
+    out.push(withProxyName(cityTarget, freshProxyName(cityTarget.proxyName, attemptIndex)));
+  }
+  const fallbackTarget = stateTarget(target);
+  out.push(withProxyName(fallbackTarget, freshProxyName(fallbackTarget.proxyName, attemptIndex)));
+  return out;
+}
+async function createFreshLocationProxy(kernel, options, target) {
+  const createErrors = [];
+  for (const candidate of freshTargetCandidates(target, options.proxyZip, options.attemptIndex)) {
+    try {
+      const created = await kernel.proxies.create({
+        type: "residential",
+        name: candidate.proxyName,
+        config: candidate.level === "zip" ? { country: candidate.country, zip: candidate.zip } : candidate.config
+      });
+      if (created.id) {
+        return resolution("location_created", options.proxyMode, created.id, candidate, null, true);
+      }
+      createErrors.push(`${candidate.proxyName}: Kernel did not return a proxy id`);
+    } catch (err) {
+      createErrors.push(`${candidate.proxyName}: ${errorText2(err)}`);
+    }
+  }
+  if (process.env.KERNEL_PROXY_TYPE !== "residential") {
+    const ispTarget = stateTarget(target);
+    try {
+      const created = await kernel.proxies.create({
+        type: "isp",
+        name: freshProxyName(`mcp-serp-isp-${ispTarget.country.toLowerCase()}-${(ispTarget.state ?? "").toLowerCase()}`, options.attemptIndex),
+        config: { country: ispTarget.country, ...ispTarget.state ? { state: ispTarget.state } : {} }
+      });
+      if (created.id) {
+        return resolution("isp_created", options.proxyMode, created.id, ispTarget, null, true);
+      }
+      createErrors.push("isp: Kernel did not return a proxy id");
+    } catch (err) {
+      createErrors.push(`isp: ${errorText2(err)}`);
+    }
+  }
+  return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, target, createErrors.join(" | "));
+}
+async function deleteKernelProxyId(kernelApiKey, proxyId) {
+  if (!kernelApiKey || !proxyId) return;
+  const kernel = new import_sdk6.default({ apiKey: kernelApiKey });
+  await kernel.proxies.delete(proxyId);
+}
+async function resolveKernelProxyId(options) {
+  if (options.proxyMode === "none") {
+    return resolution("disabled", options.proxyMode, void 0, null, null);
+  }
+  if (options.proxyMode === "configured") {
+    return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, null, null);
+  }
+  const target = parseKernelLocationProxyTarget(options.location, options.gl);
+  if (!target || !options.kernelApiKey) {
+    return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, target, target ? null : "location could not be normalized to a US city/state proxy target");
+  }
+  const kernel = new import_sdk6.default({ apiKey: options.kernelApiKey });
+  try {
+    const attemptIndex = options.attemptIndex ?? 0;
+    if (options.fresh) {
+      return await createFreshLocationProxy(kernel, options, target);
+    }
+    if (attemptIndex >= 1) {
+      const escalatedTarget = escalatedTargetLevel(target, attemptIndex);
+      const createErrors2 = [];
+      try {
+        const created = await kernel.proxies.create({
+          type: "residential",
+          name: escalatedTarget.proxyName,
+          config: escalatedTarget.config
+        });
+        if (created.id) {
+          return resolution("location_created", options.proxyMode, created.id, escalatedTarget, null);
+        }
+        createErrors2.push(`${escalatedTarget.state}: Kernel did not return a proxy id`);
+      } catch (err) {
+        createErrors2.push(`${escalatedTarget.state}: ${errorText2(err)}`);
+      }
+      return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, escalatedTarget, createErrors2.join(" | "));
+    }
+    const proxies = await kernel.proxies.list();
+    const zip = knownZipFor(target, options.proxyZip);
+    const createErrors = [];
+    if (zip) {
+      const targetZip = zipTarget(target, zip);
+      const existingZip = findExistingTargetProxy(proxies, targetZip);
+      if (existingZip?.id) {
+        return resolution("location_reused", options.proxyMode, existingZip.id, targetZip, null);
+      }
+      try {
+        const created = await kernel.proxies.create({
+          type: "residential",
+          name: targetZip.proxyName,
+          config: {
+            country: targetZip.country,
+            zip
+          }
+        });
+        if (created.id) {
+          return resolution("location_created", options.proxyMode, created.id, targetZip, null);
+        }
+        createErrors.push(`${zip}: Kernel did not return a proxy id`);
+      } catch (err) {
+        createErrors.push(`${zip}: ${errorText2(err)}`);
+      }
+    }
+    const existing = findExistingProxy(proxies, target);
+    if (existing?.id) {
+      return resolution("location_reused", options.proxyMode, existing.id, target, createErrors.join(" | ") || null);
+    }
+    for (const city of target.cityCandidates) {
+      try {
+        const created = await kernel.proxies.create({
+          type: "residential",
+          name: proxyName(target.country, target.state, city),
+          config: {
+            country: target.country,
+            state: target.state,
+            city
+          }
+        });
+        if (created.id) {
+          return resolution("location_created", options.proxyMode, created.id, {
+            ...target,
+            level: "city",
+            city,
+            proxyName: proxyName(target.country, target.state, city),
+            config: {
+              country: target.country,
+              state: target.state,
+              city
+            }
+          }, null);
+        }
+        createErrors.push(`${city}: Kernel did not return a proxy id`);
+      } catch (err) {
+        createErrors.push(`${city}: ${errorText2(err)}`);
+      }
+    }
+    const fallbackTarget = stateTarget(target);
+    const existingState = findExistingStateProxy(proxies, fallbackTarget);
+    if (existingState?.id) {
+      return resolution("location_reused", options.proxyMode, existingState.id, fallbackTarget, createErrors.join(" | "));
+    }
+    try {
+      const created = await kernel.proxies.create({
+        type: "residential",
+        name: fallbackTarget.proxyName,
+        config: fallbackTarget.config
+      });
+      if (created.id) {
+        return resolution("location_created", options.proxyMode, created.id, fallbackTarget, createErrors.join(" | "));
+      }
+      createErrors.push(`${fallbackTarget.state}: Kernel did not return a proxy id`);
+    } catch (err) {
+      createErrors.push(`${fallbackTarget.state}: ${errorText2(err)}`);
+    }
+    return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, target, createErrors.join(" | "));
+  } catch (err) {
+    return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, target, errorText2(err));
+  }
+}
+var import_sdk6, US_STATE_CODES, US_CITY_CENTER_ZIPS;
+var init_kernel_proxy_resolver = __esm({
+  "src/kernel-proxy-resolver.ts"() {
+    "use strict";
+    import_sdk6 = __toESM(require("@onkernel/sdk"), 1);
+    init_uule();
+    US_STATE_CODES = {
+      alabama: "AL",
+      alaska: "AK",
+      arizona: "AZ",
+      arkansas: "AR",
+      california: "CA",
+      colorado: "CO",
+      connecticut: "CT",
+      delaware: "DE",
+      florida: "FL",
+      georgia: "GA",
+      hawaii: "HI",
+      idaho: "ID",
+      illinois: "IL",
+      indiana: "IN",
+      iowa: "IA",
+      kansas: "KS",
+      kentucky: "KY",
+      louisiana: "LA",
+      maine: "ME",
+      maryland: "MD",
+      massachusetts: "MA",
+      michigan: "MI",
+      minnesota: "MN",
+      mississippi: "MS",
+      missouri: "MO",
+      montana: "MT",
+      nebraska: "NE",
+      nevada: "NV",
+      "new hampshire": "NH",
+      "new jersey": "NJ",
+      "new mexico": "NM",
+      "new york": "NY",
+      "north carolina": "NC",
+      "north dakota": "ND",
+      ohio: "OH",
+      oklahoma: "OK",
+      oregon: "OR",
+      pennsylvania: "PA",
+      "rhode island": "RI",
+      "south carolina": "SC",
+      "south dakota": "SD",
+      tennessee: "TN",
+      texas: "TX",
+      utah: "UT",
+      vermont: "VT",
+      virginia: "VA",
+      washington: "WA",
+      "west virginia": "WV",
+      wisconsin: "WI",
+      wyoming: "WY"
+    };
+    US_CITY_CENTER_ZIPS = {
+      "atlanta|GA": "30303",
+      "austin|TX": "78701",
+      "baltimore|MD": "21201",
+      "boston|MA": "02108",
+      "boulder|CO": "80302",
+      "charlotte|NC": "28202",
+      "chicago|IL": "60601",
+      "colorado_springs|CO": "80903",
+      "columbus|OH": "43215",
+      "dallas|TX": "75201",
+      "denver|CO": "80202",
+      "detroit|MI": "48226",
+      "fort_collins|CO": "80524",
+      "fort_worth|TX": "76102",
+      "houston|TX": "77002",
+      "indianapolis|IN": "46204",
+      "jacksonville|FL": "32202",
+      "las_vegas|NV": "89101",
+      "los_angeles|CA": "90012",
+      "louisville|KY": "40202",
+      "loveland|CO": "80537",
+      "memphis|TN": "38103",
+      "miami|FL": "33131",
+      "minneapolis|MN": "55401",
+      "nashville|TN": "37203",
+      "new_york|NY": "10001",
+      "orlando|FL": "32801",
+      "philadelphia|PA": "19103",
+      "phoenix|AZ": "85004",
+      "portland|OR": "97205",
+      "raleigh|NC": "27601",
+      "richmond|VA": "23219",
+      "sacramento|CA": "95814",
+      "salt_lake_city|UT": "84101",
+      "san_antonio|TX": "78205",
+      "san_diego|CA": "92101",
+      "san_francisco|CA": "94103",
+      "san_jose|CA": "95113",
+      "seattle|WA": "98101"
+    };
+  }
+});
+
 // src/youtube/CaptionFetcher.ts
 async function fetchViaYoutubeTranscript(videoId) {
   try {
@@ -12846,6 +13549,11 @@ async function fetchViaYoutubeTranscript(videoId) {
     }));
     const text = chunks.map((c) => c.text).join(" ");
     const last = chunks[chunks.length - 1];
+    const now = Date.now();
+    await runWithCostContext(
+      { ...currentCostContext(), op: "yt_transcription", subOp: "youtube-transcript" },
+      () => recordKernelSession({ kernelSessionId: null, source: "yt_caption_tier", headlessSent: null, openedAtMs: now, closedAtMs: now })
+    );
     return { videoId, text, chunks, durationMs: last ? last.timestamp[1] * 1e3 : 0, method: "youtube-transcript" };
   } catch {
     return null;
@@ -12897,162 +13605,214 @@ function parseTimedtextXml(xml) {
   }
   return results;
 }
-async function fetchViaKernelInnertube(videoId) {
-  const kernelApiKey = browserServiceApiKey();
-  if (!kernelApiKey) return null;
-  const kernelProxyId = browserServiceProxyId();
-  const driver = new BrowserDriver();
-  const start = Date.now();
-  try {
-    await driver.launch({ kernelApiKey, kernelProxyId, headless: true, viewport: { width: 1280, height: 800 }, locale: "en-US" });
-    const page = driver.getPage();
-    await driver.navigateTo(`https://www.youtube.com/watch?v=${videoId}`);
-    await page.waitForFunction(
-      () => !!window.ytcfg,
-      { timeout: 2e4 }
-    );
-    const xml = await page.evaluate(async (vid) => {
-      const ANDROID_VERSION = "20.10.38";
-      const resp = await fetch("/youtubei/v1/player?prettyPrint=false", {
-        method: "POST",
-        headers: {
-          "Content-Type": "application/json",
-          "User-Agent": `com.google.android.youtube/${ANDROID_VERSION} (Linux; U; Android 14)`
-        },
-        body: JSON.stringify({
-          context: { client: { clientName: "ANDROID", clientVersion: ANDROID_VERSION } },
-          videoId: vid
-        })
+function isBotBlockStatus(status) {
+  return status === "LOGIN_REQUIRED" || status === "AGE_VERIFICATION_REQUIRED";
+}
+async function probePlayer(page, videoId) {
+  return page.evaluate(
+    async ({ vid, clients }) => {
+      let lastStatus = null;
+      for (const client2 of clients) {
+        const resp = await fetch("/youtubei/v1/player?prettyPrint=false", {
+          method: "POST",
+          headers: { "Content-Type": "application/json", "User-Agent": client2.userAgent },
+          body: JSON.stringify({
+            context: { client: { clientName: client2.name, clientVersion: client2.version } },
+            videoId: vid
+          })
+        }).catch(() => null);
+        if (!resp?.ok) {
+          lastStatus = resp ? `player_http_${resp.status}` : "player_fetch_failed";
+          continue;
+        }
+        const data = await resp.json();
+        lastStatus = data?.playabilityStatus?.status ?? null;
+        const audioFormats = (data?.streamingData?.adaptiveFormats ?? []).filter((f) => typeof f.url === "string" && f.url && typeof f.mimeType === "string" && f.mimeType.startsWith("audio/")).sort((a, b) => (Number(a.contentLength) || Infinity) - (Number(b.contentLength) || Infinity));
+        const pick = audioFormats[0] ?? null;
+        let captions = null;
+        const tracks = data?.captions?.playerCaptionsTracklistRenderer?.captionTracks ?? [];
+        if (tracks.length > 0) {
+          const track = tracks.find((t) => t.languageCode === "en") ?? tracks[0];
+          const sep = track.baseUrl.includes("?") ? "&" : "?";
+          const xmlResp = await fetch(`${track.baseUrl}${sep}fmt=json3`).catch(() => null);
+          if (xmlResp?.ok) captions = await xmlResp.text();
+        }
+        if (!captions && !pick) continue;
+        return {
+          captions,
+          status: lastStatus,
+          audio: pick ? {
+            url: pick.url,
+            mimeType: pick.mimeType,
+            bitrate: pick.bitrate ?? null,
+            contentLength: pick.contentLength ? Number(pick.contentLength) || null : null,
+            approxDurationMs: pick.approxDurationMs ? Number(pick.approxDurationMs) || null : null
+          } : null
+        };
+      }
+      return { captions: null, audio: null, status: lastStatus };
+    },
+    { vid: videoId, clients: INNERTUBE_CLIENTS }
+  ).catch(() => null);
+}
+async function downloadWholeAudioStreamInOneRequestBecauseYouTubeAllowsOnlyOnePerIp(page, audio) {
+  if (audio.contentLength && audio.contentLength > MAX_AUDIO_BYTES) {
+    console.warn(`[CaptionFetcher] audio stream ${audio.contentLength} bytes exceeds ${MAX_AUDIO_BYTES} byte cap`);
+    return null;
+  }
+  const sep = audio.url.includes("?") ? "&" : "?";
+  const url = audio.contentLength ? `${audio.url}${sep}range=0-${audio.contentLength - 1}` : audio.url;
+  const result = await page.evaluate(async (u) => {
+    try {
+      const resp = await fetch(u);
+      if (!resp.ok) return { ok: false, error: `http_${resp.status}` };
+      const blob = await resp.blob();
+      const dataUrl = await new Promise((resolve, reject) => {
+        const reader = new FileReader();
+        reader.onload = () => resolve(reader.result);
+        reader.onerror = () => reject(reader.error);
+        reader.readAsDataURL(blob);
       });
-      if (!resp.ok) return null;
-      const data = await resp.json();
-      const tracks = data?.captions?.playerCaptionsTracklistRenderer?.captionTracks ?? [];
-      if (tracks.length === 0) return null;
-      const track = tracks.find((t) => t.languageCode === "en") ?? tracks[0];
-      const sep = track.baseUrl.includes("?") ? "&" : "?";
-      const xmlResp = await fetch(`${track.baseUrl}${sep}fmt=json3`);
-      return xmlResp.ok ? xmlResp.text() : null;
-    }, videoId);
-    if (!xml) return null;
-    let entries = parseTimedtextJson3(xml);
-    if (entries.length === 0) entries = parseTimedtextXml(xml);
-    if (entries.length === 0) return null;
-    const chunks = entries.map((e) => ({
-      timestamp: [e.start / 1e3, (e.start + e.dur) / 1e3],
-      text: e.text
-    }));
-    const text = chunks.map((c) => c.text).join(" ");
-    const last = chunks[chunks.length - 1];
-    return { videoId, text, chunks, durationMs: Date.now() - start, method: "browser-innertube" };
-  } catch {
+      return { ok: true, base64: dataUrl.slice(dataUrl.indexOf(",") + 1) };
+    } catch (e) {
+      return { ok: false, error: String(e) };
+    }
+  }, url).catch((e) => ({ ok: false, error: String(e) }));
+  if (!result.ok || !result.base64) {
+    console.warn(`[CaptionFetcher] audio fetch failed (${result.error ?? "unknown"}) for a ${audio.contentLength ?? 0} byte stream`);
+    return null;
+  }
+  const bytes = Buffer.from(result.base64, "base64");
+  if (audio.contentLength && bytes.length < audio.contentLength) {
+    console.warn(`[CaptionFetcher] partial audio: ${bytes.length} of ${audio.contentLength} bytes`);
     return null;
-  } finally {
-    await driver.close();
   }
+  return { bytes, mimeType: audio.mimeType.split(";")[0].trim() };
 }
-async function captureAudioViaMediaRecorder(page) {
-  const result = await page.evaluate((recordSecs) => {
-    return new Promise((resolve) => {
-      const video = document.querySelector("video");
-      if (!video) {
-        resolve({ ok: false, error: "no video element" });
-        return;
-      }
-      video.muted = false;
-      video.volume = 1;
-      try {
-        const ctx = new AudioContext();
-        const src = ctx.createMediaElementSource(video);
-        const dest = ctx.createMediaStreamDestination();
-        src.connect(dest);
-        src.connect(ctx.destination);
-        const mimeType = MediaRecorder.isTypeSupported("audio/webm;codecs=opus") ? "audio/webm;codecs=opus" : MediaRecorder.isTypeSupported("audio/webm") ? "audio/webm" : "audio/ogg";
-        const recorder = new MediaRecorder(dest.stream, { mimeType, audioBitsPerSecond: 48e3 });
-        const chunks = [];
-        recorder.ondataavailable = (e) => {
-          if (e.data.size > 0) chunks.push(e.data);
-        };
-        recorder.onstop = async () => {
-          const blob = new Blob(chunks, { type: mimeType });
-          const ab = await blob.arrayBuffer();
-          const arr = new Uint8Array(ab);
-          let binary = "";
-          const chunkSize = 8192;
-          for (let i = 0; i < arr.length; i += chunkSize) {
-            binary += String.fromCharCode(...arr.slice(i, Math.min(i + chunkSize, arr.length)));
-          }
-          resolve({ ok: true, base64: btoa(binary), size: ab.byteLength, mimeType });
-        };
-        recorder.start(1e3);
-        video.playbackRate = 2;
-        video.play().catch(() => void 0);
-        setTimeout(() => {
-          recorder.stop();
-          ctx.close();
-        }, recordSecs * 1e3);
-      } catch (e) {
-        resolve({ ok: false, error: String(e) });
-      }
-    });
-  }, WHISPER_RECORD_SECONDS);
-  if (!result.ok || !result.base64 || !result.mimeType) return null;
-  return { bytes: Buffer.from(result.base64, "base64"), mimeType: result.mimeType };
-}
-async function attemptKernelWhisper(videoId, kernelApiKey, kernelProxyId, falKey, start) {
-  const driver = new BrowserDriver();
+async function transcribeWithWizper(videoId, audio, startMs) {
+  const falKey = process.env.FAL_KEY;
+  if (!falKey) return null;
   try {
-    await driver.launch({ kernelApiKey, kernelProxyId, headless: true, viewport: { width: 1280, height: 800 }, locale: "en-US" });
-    const page = driver.getPage();
-    await driver.navigateTo(`https://www.youtube.com/watch?v=${videoId}`);
-    await page.waitForFunction(
-      () => !!window.ytcfg,
-      { timeout: 2e4 }
-    );
-    const playStatus = await page.evaluate(() => {
-      const w = window;
-      return w.ytInitialPlayerResponse?.playabilityStatus?.status;
-    });
-    if (playStatus !== "OK") {
-      await driver.close();
-      return null;
-    }
-    const audioInfo = await captureAudioViaMediaRecorder(page);
-    await driver.close();
-    if (!audioInfo || audioInfo.bytes.length < 1e4) return null;
     import_client3.fal.config({ credentials: falKey });
-    const ext = audioInfo.mimeType.includes("ogg") ? "ogg" : "webm";
-    const audioFile = new File([new Uint8Array(audioInfo.bytes)], `audio.${ext}`, { type: audioInfo.mimeType });
+    const ext = audio.mimeType.includes("webm") ? "webm" : audio.mimeType.includes("mp4") ? "m4a" : "audio";
+    const audioFile = new File([new Uint8Array(audio.bytes)], `audio.${ext}`, { type: audio.mimeType });
     const uploadedUrl = await import_client3.fal.storage.upload(audioFile);
+    const subscribeStartMs = Date.now();
     const result = await import_client3.fal.subscribe("fal-ai/wizper", {
-      input: { audio_url: uploadedUrl, task: "transcribe", language: "en" },
+      input: { audio_url: uploadedUrl, task: "transcribe", language: "en", chunk_level: "segment" },
       logs: false,
       pollInterval: 3e3
     });
+    const subscribeDurationMs = Date.now() - subscribeStartMs;
     const data = result.data;
     const chunks = (data.chunks ?? []).map((c) => ({
       timestamp: c.timestamp,
       text: c.text.trim()
     }));
-    const audioSec = chunks.length ? chunks[chunks.length - 1]?.timestamp?.[1] ?? 0 : 0;
-    void recordVendorUsage({ vendor: "fal_wizper", model: "fal-ai/wizper", units: audioSec / 60, unitType: "audio_min" });
+    void recordVendorUsage({ vendor: "fal_wizper", model: "fal-ai/wizper", units: subscribeDurationMs / 1e3, unitType: "compute_sec" });
     const text = data.text ?? chunks.map((c) => c.text).join(" ");
     if (!text) return null;
-    return { videoId, text, chunks, durationMs: Date.now() - start, method: "browser-whisper" };
-  } catch {
-    await driver.close();
+    return { videoId, text, chunks, durationMs: Date.now() - startMs, method: "browser-whisper" };
+  } catch (err) {
+    console.warn(`[CaptionFetcher] wizper transcription failed for ${videoId}: ${err instanceof Error ? err.message : String(err)}`);
     return null;
   }
 }
-async function fetchViaKernelWhisper(videoId) {
+async function resolveResidentialProxy(kernelApiKey) {
+  try {
+    const resolved = await resolveKernelProxyId({
+      kernelApiKey,
+      proxyMode: "location",
+      location: YT_RESIDENTIAL_LOCATION,
+      gl: "us",
+      fresh: true
+    });
+    return { kernelProxyId: resolved.kernelProxyId, disposableProxyId: resolved.disposableProxyId };
+  } catch (err) {
+    console.warn(`[CaptionFetcher] residential proxy resolution failed: ${err instanceof Error ? err.message : String(err)}`);
+    return {};
+  }
+}
+async function probeAndExtract(videoId, opts) {
   const kernelApiKey = browserServiceApiKey();
-  const kernelProxyId = browserServiceProxyId();
-  const falKey = process.env.FAL_KEY;
-  if (!kernelApiKey || !falKey) return null;
+  if (!kernelApiKey) return null;
+  const kernelProxyId = opts?.kernelProxyId ?? browserServiceProxyId();
+  const driver = new BrowserDriver();
+  return runWithCostContext({ ...currentCostContext(), op: "yt_transcription", subOp: "browser-innertube" }, async () => {
+    let entries = [];
+    let audio = null;
+    let retryWithFreshIp = false;
+    try {
+      await driver.launch({ kernelApiKey, kernelProxyId, headless: true, headlessMode: "headless", viewport: { width: 1280, height: 800 }, locale: "en-US" });
+      const page = driver.getPage();
+      await driver.navigateTo(`https://www.youtube.com/watch?v=${videoId}`);
+      await page.waitForFunction(
+        () => !!window.ytcfg,
+        { timeout: 2e4 }
+      );
+      const probe = await probePlayer(page, videoId);
+      if (!probe) return null;
+      if (probe.captions && !opts?.skipCaptions) {
+        entries = parseTimedtextJson3(probe.captions);
+        if (entries.length === 0) entries = parseTimedtextXml(probe.captions);
+      }
+      if (entries.length === 0) {
+        if (!probe.audio) {
+          retryWithFreshIp = isBotBlockStatus(probe.status);
+          console.warn(`[CaptionFetcher] no captions and no audio stream for ${videoId} (playability=${probe.status ?? "unknown"}, proxied=${Boolean(kernelProxyId)})`);
+          return { entries, audio: null, retryWithFreshIp };
+        }
+        audio = await downloadWholeAudioStreamInOneRequestBecauseYouTubeAllowsOnlyOnePerIp(page, probe.audio);
+        if (!audio) return { entries, audio: null, retryWithFreshIp: true };
+      }
+    } catch (err) {
+      console.warn(`[CaptionFetcher] kernel session failed for ${videoId}: ${err instanceof Error ? err.message : String(err)}`);
+      return null;
+    } finally {
+      const servedSubOp = audio ? "browser-whisper" : "browser-innertube";
+      await runWithCostContext({ ...currentCostContext(), op: "yt_transcription", subOp: servedSubOp }, () => driver.close());
+    }
+    return { entries, audio, retryWithFreshIp };
+  });
+}
+async function extractWithProxyEscalation(videoId, opts) {
+  const direct = await probeAndExtract(videoId, opts);
+  if (direct && (direct.entries.length > 0 || direct.audio)) return direct;
+  const shouldEscalate = direct === null || direct.retryWithFreshIp;
+  if (!shouldEscalate) return direct;
+  const kernelApiKey = browserServiceApiKey();
+  if (!kernelApiKey) return direct;
+  const { kernelProxyId, disposableProxyId } = await resolveResidentialProxy(kernelApiKey);
+  if (!kernelProxyId) return direct;
+  try {
+    console.warn(`[CaptionFetcher] ${videoId}: retrying through residential proxy after bot block`);
+    const viaProxy = await probeAndExtract(videoId, { ...opts, kernelProxyId });
+    return viaProxy ?? direct;
+  } finally {
+    if (disposableProxyId) {
+      await deleteKernelProxyId(kernelApiKey, disposableProxyId).catch(
+        (err) => console.warn(`[CaptionFetcher] failed to delete disposable proxy: ${err instanceof Error ? err.message : String(err)}`)
+      );
+    }
+  }
+}
+async function fetchViaKernelSession(videoId) {
   const start = Date.now();
-  for (let attempt = 0; attempt < 4; attempt++) {
-    if (attempt > 0) await new Promise((r) => setTimeout(r, 2e3 * attempt));
-    const result = await attemptKernelWhisper(videoId, kernelApiKey, kernelProxyId, falKey, start);
-    if (result) return result;
+  const outcome = await extractWithProxyEscalation(videoId);
+  if (!outcome) return null;
+  if (outcome.entries.length > 0) {
+    const chunks = outcome.entries.map((e) => ({
+      timestamp: [e.start / 1e3, (e.start + e.dur) / 1e3],
+      text: e.text
+    }));
+    const text = chunks.map((c) => c.text).join(" ");
+    return { videoId, text, chunks, durationMs: Date.now() - start, method: "browser-innertube" };
+  }
+  if (outcome.audio) {
+    return runWithCostContext(
+      { ...currentCostContext(), op: "yt_transcription", subOp: "browser-whisper" },
+      () => transcribeWithWizper(videoId, outcome.audio, start)
+    );
   }
   return null;
 }
@@ -13074,21 +13834,29 @@ function finalizeCaptions(result) {
 async function fetchCaptions(videoId) {
   const primary = await fetchViaYoutubeTranscript(videoId);
   if (primary) return finalizeCaptions(primary);
-  const kernel = await fetchViaKernelInnertube(videoId);
-  if (kernel) return finalizeCaptions(kernel);
-  const whisper = await fetchViaKernelWhisper(videoId);
-  if (whisper) return finalizeCaptions(whisper);
-  throw new Error(`No captions available for ${videoId} \u2014 all three tiers failed`);
+  const result = await fetchViaKernelSession(videoId);
+  if (result) return finalizeCaptions(result);
+  throw new Error(
+    `Could not transcribe ${videoId}. It has no caption track, and its audio stream could not be downloaded \u2014 YouTube permits a single media request per IP, so streams that do not fit in one response are unreachable. Captioned videos of any length still work.`
+  );
 }
-var import_client3, WHISPER_RECORD_SECONDS;
+var import_client3, INNERTUBE_CLIENTS, MAX_AUDIO_BYTES, YT_RESIDENTIAL_LOCATION;
 var init_CaptionFetcher = __esm({
   "src/youtube/CaptionFetcher.ts"() {
     "use strict";
     init_BrowserDriver();
     init_browser_service_env();
+    init_kernel_proxy_resolver();
     import_client3 = require("@fal-ai/client");
     init_cost_telemetry();
-    WHISPER_RECORD_SECONDS = 90;
+    init_cost_context();
+    INNERTUBE_CLIENTS = [
+      { name: "ANDROID", version: "20.10.38", userAgent: "com.google.android.youtube/20.10.38 (Linux; U; Android 14)" },
+      { name: "IOS", version: "20.10.4", userAgent: "com.google.ios.youtube/20.10.4 (iPhone16,2; U; CPU iOS 18_3_2 like Mac OS X;)" },
+      { name: "ANDROID_VR", version: "1.62.27", userAgent: "com.google.android.apps.youtube.vr.oculus/1.62.27 (Linux; U; Android 12L) gzip" }
+    ];
+    MAX_AUDIO_BYTES = 96 * 1024 * 1024;
+    YT_RESIDENTIAL_LOCATION = process.env.YOUTUBE_PROXY_LOCATION ?? "Austin, Texas";
   }
 });
 
@@ -13113,7 +13881,7 @@ var init_api_auth = __esm({
 });
 
 // src/api/server-schemas.ts
-var import_zod13, HarvestBodySchema, ExtractUrlBodySchema, MapUrlsBodySchema, ExtractSiteBodySchema, YoutubeHarvestBodySchema, YoutubeTranscribeBodySchema;
+var import_zod13, HarvestBodySchema, ExtractUrlBodySchema, DiffPageBodySchema, MapUrlsBodySchema, ExtractSiteBodySchema, YoutubeHarvestBodySchema, YoutubeTranscribeBodySchema;
 var init_server_schemas = __esm({
   "src/api/server-schemas.ts"() {
     "use strict";
@@ -13144,6 +13912,11 @@ var init_server_schemas = __esm({
       depositToVault: import_zod13.z.boolean().optional(),
       vaultName: import_zod13.z.string().trim().min(1).max(120).optional()
     });
+    DiffPageBodySchema = import_zod13.z.object({
+      url: import_zod13.z.string().min(1, "url is required"),
+      allowLocal: import_zod13.z.boolean().optional(),
+      resetBaseline: import_zod13.z.boolean().optional()
+    });
     MapUrlsBodySchema = import_zod13.z.object({
       url: import_zod13.z.string().min(1, "url is required"),
       maxUrls: import_zod13.z.number().int().min(1).max(2e3).optional(),
@@ -13419,6 +14192,7 @@ var init_youtube_routes = __esm({
     init_api_auth();
     init_server_schemas();
     init_concurrency_gates();
+    init_cost_context();
     youtubeApp = new import_hono3.Hono();
     youtubeApp.post("/harvest", createApiKeyAuth(), async (c) => {
       const raw = await c.req.json().catch(() => ({}));
@@ -13496,7 +14270,7 @@ var init_youtube_routes = __esm({
         const { ok: trOk, balance_mc: trBal } = await debitMc(user.id, holdMc, LedgerOperation.TRANSCRIPTION_HOLD, id);
         if (!trOk) return c.json(insufficientBalanceResponse(trBal, holdMc), 402);
         debited = true;
-        const result = await fetchCaptions(id);
+        const result = await runWithCostContext({ ...currentCostContext(), op: "yt_transcription", userId: user.id }, () => fetchCaptions(id));
         const lastChunk = result.chunks?.[result.chunks.length - 1];
         const chunkSecs = lastChunk && Number.isFinite(lastChunk.timestamp[1]) ? lastChunk.timestamp[1] : 0;
         const durationSecs = Number.isFinite(result.durationMs) ? result.durationMs / 1e3 : 0;
@@ -14317,755 +15091,198 @@ function extractCollatedResults(payload) {
         page_name: r.page_name ?? snapshot?.page_name ?? "",
         is_active: Boolean(r.is_active),
         collation_count: typeof r.collation_count === "number" ? r.collation_count : null,
-        snapshot
-      });
-    }
-  }
-  return results;
-}
-async function collectAdLibraryResults(page, url, maxResults, opts = {}) {
-  const captureMs = opts.captureMs ?? 3e4;
-  const collected = [];
-  const seen = /* @__PURE__ */ new Set();
-  const handler = (resp) => {
-    if (!resp.url().includes("/api/graphql")) return;
-    const friendlyName = (resp.request().postData() ?? "").match(/fb_api_req_friendly_name=([^&]+)/)?.[1];
-    if (friendlyName !== AD_LIBRARY_QUERY) return;
-    void resp.text().then((text) => {
-      for (const payload of parseFbGraphqlJson(text)) {
-        for (const result of extractCollatedResults(payload)) {
-          if (seen.has(result.ad_archive_id)) continue;
-          seen.add(result.ad_archive_id);
-          collected.push(result);
-        }
-      }
-    }).catch(() => void 0);
-  };
-  page.on("response", handler);
-  try {
-    await page.goto(url, { waitUntil: "domcontentloaded", timeout: 45e3 });
-    const deadline = Date.now() + captureMs;
-    let lastCount = -1;
-    let stableRounds = 0;
-    while (Date.now() < deadline && collected.length < maxResults) {
-      await page.evaluate(() => window.scrollTo(0, document.body.scrollHeight)).catch(() => void 0);
-      await page.waitForTimeout(2e3);
-      if (collected.length === lastCount) {
-        stableRounds++;
-        if (stableRounds >= 2 && collected.length > 0) break;
-      } else {
-        stableRounds = 0;
-      }
-      lastCount = collected.length;
-    }
-  } finally {
-    page.off("response", handler);
-  }
-  return collected.slice(0, maxResults);
-}
-function advertisersFromResults(results, maxResults) {
-  const byPage = /* @__PURE__ */ new Map();
-  for (const r of results) {
-    if (!r.page_id || !r.page_name) continue;
-    const collation = typeof r.collation_count === "number" && r.collation_count > 0 ? r.collation_count : 0;
-    const existing = byPage.get(r.page_id);
-    if (existing) {
-      existing.resultCount++;
-      existing.maxCollation = Math.max(existing.maxCollation, collation);
-    } else {
-      byPage.set(r.page_id, { pageName: r.page_name, pageId: r.page_id, sampleLibraryId: r.ad_archive_id, maxCollation: collation, resultCount: 1 });
-    }
-  }
-  return [...byPage.values()].map((e) => ({ pageName: e.pageName, pageId: e.pageId, sampleLibraryId: e.sampleLibraryId, adCount: Math.max(e.maxCollation, e.resultCount) })).sort((a, b) => b.adCount - a.adCount).slice(0, maxResults);
-}
-var AD_LIBRARY_QUERY;
-var init_FacebookAdGraphql = __esm({
-  "src/extractor/FacebookAdGraphql.ts"() {
-    "use strict";
-    AD_LIBRARY_QUERY = "AdLibrarySearchPaginationQuery";
-  }
-});
-
-// src/extractor/FacebookOrganicVideoExtractor.ts
-function htmlDecode(value) {
-  return value.replace(/&amp;/g, "&").replace(/&quot;/g, '"').replace(/&#039;/g, "'").replace(/&lt;/g, "<").replace(/&gt;/g, ">");
-}
-function decodeEscapedValue(raw) {
-  let out = raw;
-  try {
-    out = JSON.parse(`"${raw.replace(/"/g, '\\"')}"`);
-  } catch {
-    out = raw.replace(/\\u0025/g, "%").replace(/\\u0026/g, "&").replace(/\\u003D/g, "=").replace(/\\u003d/g, "=").replace(/\\"/g, '"').replace(/\\\//g, "/");
-  }
-  return htmlDecode(out);
-}
-function facebookVideoIdFromUrl(url) {
-  try {
-    const parsed = new URL(url);
-    const path6 = parsed.pathname;
-    const direct = path6.match(/\/(?:reel|videos|watch)\/(?:[^/]+\/)?(\d{8,})/i);
-    if (direct) return direct[1];
-    const queryVideoId = parsed.searchParams.get("v") ?? parsed.searchParams.get("video_id");
-    if (queryVideoId && /^\d{8,}$/.test(queryVideoId)) return queryVideoId;
-    const numericTail = path6.match(/\/(\d{8,})\/?$/);
-    if (numericTail) return numericTail[1];
-  } catch {
-  }
-  return null;
-}
-function parseEfg(url) {
-  try {
-    const efg = new URL(url).searchParams.get("efg");
-    if (!efg) return null;
-    return JSON.parse(Buffer.from(efg, "base64").toString("utf8"));
-  } catch {
-    return null;
-  }
-}
-function manifestDurationSec(htmlWindow) {
-  const decoded = decodeEscapedValue(htmlWindow);
-  const match = decoded.match(/mediaPresentationDuration="PT([\d.]+)S"/);
-  return match ? Number(match[1]) : null;
-}
-function metadataContent(html, key) {
-  const escaped = key.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
-  const re = new RegExp(`<meta\\s+[^>]*(?:property|name)=["']${escaped}["'][^>]*content=["']([^"']+)["'][^>]*>`, "i");
-  const m = html.match(re);
-  return m ? htmlDecode(m[1]).trim() : null;
-}
-function ownerNameFromHtml(html) {
-  const ogTitle = metadataContent(html, "og:title");
-  const pipe = ogTitle?.split("|").map((s) => s.trim()).filter(Boolean);
-  if (pipe && pipe.length > 1) return pipe[pipe.length - 1] ?? null;
-  return null;
-}
-function extractFacebookOrganicVideo(html, pageUrl, sourceUrl = pageUrl, quality = "best") {
-  const videoId = facebookVideoIdFromUrl(pageUrl) ?? facebookVideoIdFromUrl(sourceUrl);
-  const manifestIndex = videoId ? html.indexOf(`dash_mpd_debug.mpd?v=${videoId}`) : -1;
-  const searchWindow = manifestIndex >= 0 ? html.slice(Math.max(0, manifestIndex - 5e4), Math.min(html.length, manifestIndex + 5e4)) : html;
-  const targetDuration = manifestDurationSec(searchWindow);
-  const candidates = [];
-  const seen = /* @__PURE__ */ new Set();
-  const re = /browser_native_(sd|hd)_url\\?"\s*:\s*\\?"([^"<]+?)\\?"/g;
-  let match;
-  while ((match = re.exec(searchWindow)) !== null) {
-    const url = decodeEscapedValue(match[2]);
-    if (!url || seen.has(url)) continue;
-    seen.add(url);
-    const parsed = parseEfg(url);
-    let bitrate = null;
-    try {
-      const value = new URL(url).searchParams.get("bitrate");
-      bitrate = value ? Number(value) : null;
-    } catch {
-    }
-    candidates.push({
-      url,
-      quality: match[1],
-      bitrate: Number.isFinite(bitrate) ? bitrate : null,
-      durationSec: typeof parsed?.duration_s === "number" ? parsed.duration_s : null,
-      assetId: typeof parsed?.xpv_asset_id === "number" ? parsed.xpv_asset_id : null,
-      vencodeTag: typeof parsed?.vencode_tag === "string" ? parsed.vencode_tag : null
-    });
-  }
-  const targetCandidates = candidates.filter((candidate) => {
-    if (!targetDuration || candidate.durationSec == null) return true;
-    return Math.abs(candidate.durationSec - targetDuration) <= 2;
-  });
-  const selectable = targetCandidates.length ? targetCandidates : candidates;
-  const selected = [...selectable].filter((candidate) => quality === "best" || candidate.quality === quality).sort((a, b) => {
-    const aq = a.quality === "hd" ? 1 : 0;
-    const bq = b.quality === "hd" ? 1 : 0;
-    if (quality === "best" && aq !== bq) return bq - aq;
-    return (b.bitrate ?? 0) - (a.bitrate ?? 0);
-  })[0];
-  if (!selected) {
-    throw new Error("No progressive Facebook video URL was found in the rendered page state");
-  }
-  return {
-    sourceUrl,
-    pageUrl,
-    videoId,
-    title: metadataContent(html, "og:title") ?? metadataContent(html, "twitter:title"),
-    description: metadataContent(html, "og:description") ?? metadataContent(html, "description") ?? metadataContent(html, "twitter:description"),
-    ownerName: ownerNameFromHtml(html),
-    videoUrl: selected.url,
-    selectedQuality: selected.quality,
-    bitrate: selected.bitrate,
-    durationSec: targetDuration ?? selected.durationSec,
-    assetId: selected.assetId,
-    vencodeTag: selected.vencodeTag,
-    candidates: selectable,
-    extractedAt: (/* @__PURE__ */ new Date()).toISOString()
-  };
-}
-async function extractFacebookOrganicVideoFromPage(page, sourceUrl, quality = "best") {
-  await page.waitForLoadState("domcontentloaded", { timeout: 15e3 }).catch(() => {
-  });
-  await page.waitForTimeout(1500);
-  const html = await page.content();
-  return extractFacebookOrganicVideo(html, page.url(), sourceUrl, quality);
-}
-var init_FacebookOrganicVideoExtractor = __esm({
-  "src/extractor/FacebookOrganicVideoExtractor.ts"() {
-    "use strict";
-  }
-});
-
-// src/locations.ts
-var LOCATIONS;
-var init_locations = __esm({
-  "src/locations.ts"() {
-    "use strict";
-    LOCATIONS = {
-      "austin": "Austin,Texas,United States",
-      "new york": "New York,New York,United States",
-      "new york city": "New York,New York,United States",
-      "nyc": "New York,New York,United States",
-      "los angeles": "Los Angeles,California,United States",
-      "la": "Los Angeles,California,United States",
-      "chicago": "Chicago,Illinois,United States",
-      "houston": "Houston,Texas,United States",
-      "phoenix": "Phoenix,Arizona,United States",
-      "philadelphia": "Philadelphia,Pennsylvania,United States",
-      "philly": "Philadelphia,Pennsylvania,United States",
-      "san antonio": "San Antonio,Texas,United States",
-      "dallas": "Dallas,Texas,United States",
-      "miami": "Miami,Florida,United States",
-      "seattle": "Seattle,Washington,United States",
-      "denver": "Denver,Colorado,United States",
-      "loveland": "Loveland,Colorado,United States",
-      "loveland co": "Loveland,Colorado,United States",
-      "fort collins": "Fort Collins,Colorado,United States",
-      "boulder": "Boulder,Colorado,United States",
-      "colorado springs": "Colorado Springs,Colorado,United States",
-      "boston": "Boston,Massachusetts,United States",
-      "atlanta": "Atlanta,Georgia,United States",
-      "san francisco": "San Francisco,California,United States",
-      "sf": "San Francisco,California,United States",
-      "portland": "Portland,Oregon,United States",
-      "las vegas": "Las Vegas,Nevada,United States",
-      "minneapolis": "Minneapolis,Minnesota,United States",
-      "detroit": "Detroit,Michigan,United States",
-      "nashville": "Nashville,Tennessee,United States",
-      "charlotte": "Charlotte,North Carolina,United States",
-      "orlando": "Orlando,Florida,United States",
-      "san diego": "San Diego,California,United States",
-      "baltimore": "Baltimore,Maryland,United States",
-      "sacramento": "Sacramento,California,United States",
-      "columbus": "Columbus,Ohio,United States",
-      "indianapolis": "Indianapolis,Indiana,United States",
-      "san jose": "San Jose,California,United States",
-      "fort worth": "Fort Worth,Texas,United States",
-      "jacksonville": "Jacksonville,Florida,United States",
-      "memphis": "Memphis,Tennessee,United States",
-      "louisville": "Louisville,Kentucky,United States",
-      "raleigh": "Raleigh,North Carolina,United States",
-      "richmond": "Richmond,Virginia,United States",
-      "salt lake city": "Salt Lake City,Utah,United States",
-      "toronto": "Toronto,Ontario,Canada",
-      "vancouver": "Vancouver,British Columbia,Canada",
-      "montreal": "Montreal,Quebec,Canada",
-      "calgary": "Calgary,Alberta,Canada",
-      "ottawa": "Ottawa,Ontario,Canada",
-      "london": "London,England,United Kingdom",
-      "manchester": "Manchester,England,United Kingdom",
-      "birmingham": "Birmingham,England,United Kingdom",
-      "edinburgh": "Edinburgh,Scotland,United Kingdom",
-      "glasgow": "Glasgow,Scotland,United Kingdom",
-      "leeds": "Leeds,England,United Kingdom",
-      "sydney": "Sydney,New South Wales,Australia",
-      "melbourne": "Melbourne,Victoria,Australia",
-      "brisbane": "Brisbane,Queensland,Australia",
-      "perth": "Perth,Western Australia,Australia",
-      "adelaide": "Adelaide,South Australia,Australia",
-      "dublin": "Dublin,Leinster,Ireland"
-    };
-  }
-});
-
-// src/uule.ts
-function encodeVarint(value) {
-  const bytes = [];
-  let remaining = value;
-  do {
-    let byte = remaining & 127;
-    remaining >>>= 7;
-    if (remaining > 0) byte |= 128;
-    bytes.push(byte);
-  } while (remaining > 0);
-  return bytes;
-}
-function encodeUule(name) {
-  const locationBytes = Buffer.from(name, "utf8");
-  const payload = Buffer.concat([
-    Buffer.from([8, 2, 16, 32, 34]),
-    Buffer.from(encodeVarint(locationBytes.length)),
-    locationBytes
-  ]);
-  return `w+${payload.toString("base64")}`;
-}
-function normalizeLocation(input) {
-  const raw = input.toLowerCase().trim();
-  if (LOCATIONS[raw]) return LOCATIONS[raw];
-  const beforeComma = raw.split(",")[0].trim();
-  if (beforeComma !== raw && LOCATIONS[beforeComma]) return LOCATIONS[beforeComma];
-  const withoutState = raw.replace(/\s+[a-z]{2}$/, "").trim();
-  if (withoutState !== raw && LOCATIONS[withoutState]) return LOCATIONS[withoutState];
-  return input;
-}
-var init_uule = __esm({
-  "src/uule.ts"() {
-    "use strict";
-    init_locations();
-  }
-});
-
-// src/kernel-proxy-resolver.ts
-function proxyIdSuffix2(proxyId) {
-  return proxyId ? proxyId.slice(-6) : null;
-}
-function resolution(source, proxyMode, proxyId, target, error, disposable = false) {
-  const proxyType = source === "disabled" ? "none" : source === "configured_fallback" ? "configured" : source === "isp_created" ? "isp" : source === "location_created" || source === "location_reused" ? "residential" : "unknown";
-  return {
-    kernelProxyId: proxyId,
-    ...disposable && proxyId ? { disposableProxyId: proxyId } : {},
-    resolution: {
-      source,
-      proxyType,
-      proxyMode,
-      proxyIdPresent: Boolean(proxyId),
-      proxyIdSuffix: proxyIdSuffix2(proxyId),
-      target,
-      error,
-      disposable
-    }
-  };
-}
-function normalizeStateName(value) {
-  return value.trim().toLowerCase().replace(/\s+/g, " ");
-}
-function normalizeCountryName(value) {
-  return value.trim().toLowerCase().replace(/\./g, "").replace(/\s+/g, " ");
-}
-function isUnitedStates(country) {
-  if (!country) return true;
-  const normalized = normalizeCountryName(country);
-  return normalized === "united states" || normalized === "united states of america" || normalized === "usa" || normalized === "us";
-}
-function stateCodeFor(region) {
-  const trimmed = region.trim();
-  if (/^[A-Za-z]{2}$/.test(trimmed)) return trimmed.toUpperCase();
-  return US_STATE_CODES[normalizeStateName(trimmed)] ?? null;
-}
-function kernelCityIdentifierCandidates(city) {
-  const ascii = city.normalize("NFKD").replace(/[^\x00-\x7F]/g, "").toLowerCase();
-  const words = ascii.split(/[^a-z0-9]+/).filter(Boolean);
-  const underscored = words.join("_");
-  const compact = words.join("");
-  return Array.from(new Set([underscored, compact].filter(Boolean)));
-}
-function proxyName(country, state, city) {
-  return city ? `mcp-serp-residential-${country.toLowerCase()}-${state.toLowerCase()}-${city}` : `mcp-serp-residential-${country.toLowerCase()}-${state.toLowerCase()}`;
-}
-function freshProxyName(baseName, attemptIndex) {
-  const stamp = `${Date.now()}-${attemptIndex ?? 0}-${Math.random().toString(36).slice(2, 8)}`;
-  return `${baseName}-fresh-${stamp}`;
-}
-function zipProxyName(zip) {
-  return `mcp-serp-residential-us-zip-${zip}`;
-}
-function parseKernelLocationProxyTarget(location2, gl) {
-  if (!location2 || gl.toLowerCase() !== "us") return null;
-  const canonicalLocation = normalizeLocation(location2);
-  let parts = canonicalLocation.split(",").map((part) => part.trim()).filter(Boolean);
-  if (parts.length > 1 && isUnitedStates(parts[parts.length - 1])) {
-    parts = parts.slice(0, -1);
-  }
-  if (parts.length === 1) {
-    const stateOnly = stateCodeFor(parts[0]);
-    if (!stateOnly) return null;
-    return {
-      canonicalLocation,
-      level: "state",
-      country: "US",
-      state: stateOnly,
-      city: "",
-      cityCandidates: [],
-      proxyName: proxyName("US", stateOnly),
-      config: {
-        country: "US",
-        state: stateOnly
-      }
-    };
-  }
-  const [city = "", region = ""] = parts;
-  if (!city || !region) return null;
-  const state = stateCodeFor(region);
-  if (!state) return null;
-  const cityCandidates = kernelCityIdentifierCandidates(city);
-  const primaryCity = cityCandidates[0];
-  if (!primaryCity) return null;
-  return {
-    canonicalLocation,
-    level: "city",
-    country: "US",
-    state,
-    city: primaryCity,
-    cityCandidates,
-    proxyName: proxyName("US", state, primaryCity),
-    config: {
-      country: "US",
-      state,
-      city: primaryCity
-    }
-  };
-}
-function cityZipKey(target) {
-  return `${target.city}|${target.state}`;
-}
-function knownZipFor(target, explicitZip) {
-  if (explicitZip && /^\d{5}$/.test(explicitZip)) return explicitZip;
-  return US_CITY_CENTER_ZIPS[cityZipKey(target)] ?? null;
-}
-function zipTarget(target, zip) {
-  return {
-    ...target,
-    level: "zip",
-    zip,
-    proxyName: zipProxyName(zip),
-    config: {
-      country: target.country,
-      state: target.state,
-      zip
-    }
-  };
-}
-function withProxyName(target, name) {
-  return {
-    ...target,
-    proxyName: name
-  };
-}
-function configMatches(config, target, city) {
-  if (target.level === "zip") {
-    return config?.country?.toUpperCase() === target.country && config?.zip === target.zip;
-  }
-  return config?.country?.toUpperCase() === target.country && config?.state?.toUpperCase() === target.state && (city ? config?.city === city : !config?.city);
-}
-function findExistingTargetProxy(proxies, target) {
-  return proxies.find((proxy) => proxy.type === "residential" && proxy.status !== "unavailable" && Boolean(proxy.id) && (proxy.name === target.proxyName || configMatches(proxy.config, target, target.level === "city" ? target.city : void 0))) ?? null;
-}
-function findExistingProxy(proxies, target) {
-  for (const city of target.cityCandidates) {
-    const name = proxyName(target.country, target.state, city);
-    const found = proxies.find((proxy) => proxy.type === "residential" && proxy.status !== "unavailable" && Boolean(proxy.id) && (proxy.name === name || configMatches(proxy.config, target, city)));
-    if (found) return found;
-  }
-  return null;
-}
-function stateTarget(target) {
-  return {
-    ...target,
-    level: "state",
-    proxyName: proxyName(target.country, target.state),
-    config: {
-      country: target.country,
-      state: target.state
-    }
-  };
-}
-function findExistingStateProxy(proxies, target) {
-  const name = proxyName(target.country, target.state);
-  return proxies.find((proxy) => proxy.type === "residential" && proxy.status !== "unavailable" && Boolean(proxy.id) && (proxy.name === name || configMatches(proxy.config, target))) ?? null;
-}
-function escalatedTargetLevel(target, attemptIndex) {
-  return stateTarget(target);
-}
-function errorText2(err) {
-  return err instanceof Error ? err.message : String(err);
-}
-function freshTargetCandidates(target, explicitZip, attemptIndex) {
-  const out = [];
-  const zip = knownZipFor(target, explicitZip);
-  if (zip) {
-    const targetZip = zipTarget(target, zip);
-    out.push(withProxyName(targetZip, freshProxyName(targetZip.proxyName, attemptIndex)));
-  }
-  for (const city of target.cityCandidates) {
-    const cityTarget = {
-      ...target,
-      level: "city",
-      city,
-      proxyName: proxyName(target.country, target.state, city),
-      config: {
-        country: target.country,
-        state: target.state,
-        city
-      }
-    };
-    out.push(withProxyName(cityTarget, freshProxyName(cityTarget.proxyName, attemptIndex)));
-  }
-  const fallbackTarget = stateTarget(target);
-  out.push(withProxyName(fallbackTarget, freshProxyName(fallbackTarget.proxyName, attemptIndex)));
-  return out;
-}
-async function createFreshLocationProxy(kernel, options, target) {
-  const createErrors = [];
-  for (const candidate of freshTargetCandidates(target, options.proxyZip, options.attemptIndex)) {
-    try {
-      const created = await kernel.proxies.create({
-        type: "residential",
-        name: candidate.proxyName,
-        config: candidate.level === "zip" ? { country: candidate.country, zip: candidate.zip } : candidate.config
+        snapshot
       });
-      if (created.id) {
-        return resolution("location_created", options.proxyMode, created.id, candidate, null, true);
-      }
-      createErrors.push(`${candidate.proxyName}: Kernel did not return a proxy id`);
-    } catch (err) {
-      createErrors.push(`${candidate.proxyName}: ${errorText2(err)}`);
     }
   }
-  if (process.env.KERNEL_PROXY_TYPE !== "residential") {
-    const ispTarget = stateTarget(target);
-    try {
-      const created = await kernel.proxies.create({
-        type: "isp",
-        name: freshProxyName(`mcp-serp-isp-${ispTarget.country.toLowerCase()}-${(ispTarget.state ?? "").toLowerCase()}`, options.attemptIndex),
-        config: { country: ispTarget.country, ...ispTarget.state ? { state: ispTarget.state } : {} }
-      });
-      if (created.id) {
-        return resolution("isp_created", options.proxyMode, created.id, ispTarget, null, true);
+  return results;
+}
+async function collectAdLibraryResults(page, url, maxResults, opts = {}) {
+  const captureMs = opts.captureMs ?? 3e4;
+  const collected = [];
+  const seen = /* @__PURE__ */ new Set();
+  const handler = (resp) => {
+    if (!resp.url().includes("/api/graphql")) return;
+    const friendlyName = (resp.request().postData() ?? "").match(/fb_api_req_friendly_name=([^&]+)/)?.[1];
+    if (friendlyName !== AD_LIBRARY_QUERY) return;
+    void resp.text().then((text) => {
+      for (const payload of parseFbGraphqlJson(text)) {
+        for (const result of extractCollatedResults(payload)) {
+          if (seen.has(result.ad_archive_id)) continue;
+          seen.add(result.ad_archive_id);
+          collected.push(result);
+        }
       }
-      createErrors.push("isp: Kernel did not return a proxy id");
-    } catch (err) {
-      createErrors.push(`isp: ${errorText2(err)}`);
+    }).catch(() => void 0);
+  };
+  page.on("response", handler);
+  try {
+    await page.goto(url, { waitUntil: "domcontentloaded", timeout: 45e3 });
+    const deadline = Date.now() + captureMs;
+    let lastCount = -1;
+    let stableRounds = 0;
+    while (Date.now() < deadline && collected.length < maxResults) {
+      await page.evaluate(() => window.scrollTo(0, document.body.scrollHeight)).catch(() => void 0);
+      await page.waitForTimeout(2e3);
+      if (collected.length === lastCount) {
+        stableRounds++;
+        if (stableRounds >= 2 && collected.length > 0) break;
+      } else {
+        stableRounds = 0;
+      }
+      lastCount = collected.length;
     }
+  } finally {
+    page.off("response", handler);
   }
-  return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, target, createErrors.join(" | "));
+  return collected.slice(0, maxResults);
 }
-async function deleteKernelProxyId(kernelApiKey, proxyId) {
-  if (!kernelApiKey || !proxyId) return;
-  const kernel = new import_sdk6.default({ apiKey: kernelApiKey });
-  await kernel.proxies.delete(proxyId);
+function advertisersFromResults(results, maxResults) {
+  const byPage = /* @__PURE__ */ new Map();
+  for (const r of results) {
+    if (!r.page_id || !r.page_name) continue;
+    const collation = typeof r.collation_count === "number" && r.collation_count > 0 ? r.collation_count : 0;
+    const existing = byPage.get(r.page_id);
+    if (existing) {
+      existing.resultCount++;
+      existing.maxCollation = Math.max(existing.maxCollation, collation);
+    } else {
+      byPage.set(r.page_id, { pageName: r.page_name, pageId: r.page_id, sampleLibraryId: r.ad_archive_id, maxCollation: collation, resultCount: 1 });
+    }
+  }
+  return [...byPage.values()].map((e) => ({ pageName: e.pageName, pageId: e.pageId, sampleLibraryId: e.sampleLibraryId, adCount: Math.max(e.maxCollation, e.resultCount) })).sort((a, b) => b.adCount - a.adCount).slice(0, maxResults);
 }
-async function resolveKernelProxyId(options) {
-  if (options.proxyMode === "none") {
-    return resolution("disabled", options.proxyMode, void 0, null, null);
+var AD_LIBRARY_QUERY;
+var init_FacebookAdGraphql = __esm({
+  "src/extractor/FacebookAdGraphql.ts"() {
+    "use strict";
+    AD_LIBRARY_QUERY = "AdLibrarySearchPaginationQuery";
   }
-  if (options.proxyMode === "configured") {
-    return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, null, null);
+});
+
+// src/extractor/FacebookOrganicVideoExtractor.ts
+function htmlDecode(value) {
+  return value.replace(/&amp;/g, "&").replace(/&quot;/g, '"').replace(/&#039;/g, "'").replace(/&lt;/g, "<").replace(/&gt;/g, ">");
+}
+function decodeEscapedValue(raw) {
+  let out = raw;
+  try {
+    out = JSON.parse(`"${raw.replace(/"/g, '\\"')}"`);
+  } catch {
+    out = raw.replace(/\\u0025/g, "%").replace(/\\u0026/g, "&").replace(/\\u003D/g, "=").replace(/\\u003d/g, "=").replace(/\\"/g, '"').replace(/\\\//g, "/");
   }
-  const target = parseKernelLocationProxyTarget(options.location, options.gl);
-  if (!target || !options.kernelApiKey) {
-    return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, target, target ? null : "location could not be normalized to a US city/state proxy target");
+  return htmlDecode(out);
+}
+function facebookVideoIdFromUrl(url) {
+  try {
+    const parsed = new URL(url);
+    const path6 = parsed.pathname;
+    const direct = path6.match(/\/(?:reel|videos|watch)\/(?:[^/]+\/)?(\d{8,})/i);
+    if (direct) return direct[1];
+    const queryVideoId = parsed.searchParams.get("v") ?? parsed.searchParams.get("video_id");
+    if (queryVideoId && /^\d{8,}$/.test(queryVideoId)) return queryVideoId;
+    const numericTail = path6.match(/\/(\d{8,})\/?$/);
+    if (numericTail) return numericTail[1];
+  } catch {
   }
-  const kernel = new import_sdk6.default({ apiKey: options.kernelApiKey });
+  return null;
+}
+function parseEfg(url) {
   try {
-    const attemptIndex = options.attemptIndex ?? 0;
-    if (options.fresh) {
-      return await createFreshLocationProxy(kernel, options, target);
-    }
-    if (attemptIndex >= 1) {
-      const escalatedTarget = escalatedTargetLevel(target, attemptIndex);
-      const createErrors2 = [];
-      try {
-        const created = await kernel.proxies.create({
-          type: "residential",
-          name: escalatedTarget.proxyName,
-          config: escalatedTarget.config
-        });
-        if (created.id) {
-          return resolution("location_created", options.proxyMode, created.id, escalatedTarget, null);
-        }
-        createErrors2.push(`${escalatedTarget.state}: Kernel did not return a proxy id`);
-      } catch (err) {
-        createErrors2.push(`${escalatedTarget.state}: ${errorText2(err)}`);
-      }
-      return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, escalatedTarget, createErrors2.join(" | "));
-    }
-    const proxies = await kernel.proxies.list();
-    const zip = knownZipFor(target, options.proxyZip);
-    const createErrors = [];
-    if (zip) {
-      const targetZip = zipTarget(target, zip);
-      const existingZip = findExistingTargetProxy(proxies, targetZip);
-      if (existingZip?.id) {
-        return resolution("location_reused", options.proxyMode, existingZip.id, targetZip, null);
-      }
-      try {
-        const created = await kernel.proxies.create({
-          type: "residential",
-          name: targetZip.proxyName,
-          config: {
-            country: targetZip.country,
-            zip
-          }
-        });
-        if (created.id) {
-          return resolution("location_created", options.proxyMode, created.id, targetZip, null);
-        }
-        createErrors.push(`${zip}: Kernel did not return a proxy id`);
-      } catch (err) {
-        createErrors.push(`${zip}: ${errorText2(err)}`);
-      }
-    }
-    const existing = findExistingProxy(proxies, target);
-    if (existing?.id) {
-      return resolution("location_reused", options.proxyMode, existing.id, target, createErrors.join(" | ") || null);
-    }
-    for (const city of target.cityCandidates) {
-      try {
-        const created = await kernel.proxies.create({
-          type: "residential",
-          name: proxyName(target.country, target.state, city),
-          config: {
-            country: target.country,
-            state: target.state,
-            city
-          }
-        });
-        if (created.id) {
-          return resolution("location_created", options.proxyMode, created.id, {
-            ...target,
-            level: "city",
-            city,
-            proxyName: proxyName(target.country, target.state, city),
-            config: {
-              country: target.country,
-              state: target.state,
-              city
-            }
-          }, null);
-        }
-        createErrors.push(`${city}: Kernel did not return a proxy id`);
-      } catch (err) {
-        createErrors.push(`${city}: ${errorText2(err)}`);
-      }
-    }
-    const fallbackTarget = stateTarget(target);
-    const existingState = findExistingStateProxy(proxies, fallbackTarget);
-    if (existingState?.id) {
-      return resolution("location_reused", options.proxyMode, existingState.id, fallbackTarget, createErrors.join(" | "));
-    }
+    const efg = new URL(url).searchParams.get("efg");
+    if (!efg) return null;
+    return JSON.parse(Buffer.from(efg, "base64").toString("utf8"));
+  } catch {
+    return null;
+  }
+}
+function manifestDurationSec(htmlWindow) {
+  const decoded = decodeEscapedValue(htmlWindow);
+  const match = decoded.match(/mediaPresentationDuration="PT([\d.]+)S"/);
+  return match ? Number(match[1]) : null;
+}
+function metadataContent(html, key) {
+  const escaped = key.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
+  const re = new RegExp(`<meta\\s+[^>]*(?:property|name)=["']${escaped}["'][^>]*content=["']([^"']+)["'][^>]*>`, "i");
+  const m = html.match(re);
+  return m ? htmlDecode(m[1]).trim() : null;
+}
+function ownerNameFromHtml(html) {
+  const ogTitle = metadataContent(html, "og:title");
+  const pipe = ogTitle?.split("|").map((s) => s.trim()).filter(Boolean);
+  if (pipe && pipe.length > 1) return pipe[pipe.length - 1] ?? null;
+  return null;
+}
+function extractFacebookOrganicVideo(html, pageUrl, sourceUrl = pageUrl, quality = "best") {
+  const videoId = facebookVideoIdFromUrl(pageUrl) ?? facebookVideoIdFromUrl(sourceUrl);
+  const manifestIndex = videoId ? html.indexOf(`dash_mpd_debug.mpd?v=${videoId}`) : -1;
+  const searchWindow = manifestIndex >= 0 ? html.slice(Math.max(0, manifestIndex - 5e4), Math.min(html.length, manifestIndex + 5e4)) : html;
+  const targetDuration = manifestDurationSec(searchWindow);
+  const candidates = [];
+  const seen = /* @__PURE__ */ new Set();
+  const re = /browser_native_(sd|hd)_url\\?"\s*:\s*\\?"([^"<]+?)\\?"/g;
+  let match;
+  while ((match = re.exec(searchWindow)) !== null) {
+    const url = decodeEscapedValue(match[2]);
+    if (!url || seen.has(url)) continue;
+    seen.add(url);
+    const parsed = parseEfg(url);
+    let bitrate = null;
     try {
-      const created = await kernel.proxies.create({
-        type: "residential",
-        name: fallbackTarget.proxyName,
-        config: fallbackTarget.config
-      });
-      if (created.id) {
-        return resolution("location_created", options.proxyMode, created.id, fallbackTarget, createErrors.join(" | "));
-      }
-      createErrors.push(`${fallbackTarget.state}: Kernel did not return a proxy id`);
-    } catch (err) {
-      createErrors.push(`${fallbackTarget.state}: ${errorText2(err)}`);
+      const value = new URL(url).searchParams.get("bitrate");
+      bitrate = value ? Number(value) : null;
+    } catch {
     }
-    return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, target, createErrors.join(" | "));
-  } catch (err) {
-    return resolution("configured_fallback", options.proxyMode, options.configuredKernelProxyId, target, errorText2(err));
+    candidates.push({
+      url,
+      quality: match[1],
+      bitrate: Number.isFinite(bitrate) ? bitrate : null,
+      durationSec: typeof parsed?.duration_s === "number" ? parsed.duration_s : null,
+      assetId: typeof parsed?.xpv_asset_id === "number" ? parsed.xpv_asset_id : null,
+      vencodeTag: typeof parsed?.vencode_tag === "string" ? parsed.vencode_tag : null
+    });
+  }
+  const targetCandidates = candidates.filter((candidate) => {
+    if (!targetDuration || candidate.durationSec == null) return true;
+    return Math.abs(candidate.durationSec - targetDuration) <= 2;
+  });
+  const selectable = targetCandidates.length ? targetCandidates : candidates;
+  const selected = [...selectable].filter((candidate) => quality === "best" || candidate.quality === quality).sort((a, b) => {
+    const aq = a.quality === "hd" ? 1 : 0;
+    const bq = b.quality === "hd" ? 1 : 0;
+    if (quality === "best" && aq !== bq) return bq - aq;
+    return (b.bitrate ?? 0) - (a.bitrate ?? 0);
+  })[0];
+  if (!selected) {
+    throw new Error("No progressive Facebook video URL was found in the rendered page state");
   }
+  return {
+    sourceUrl,
+    pageUrl,
+    videoId,
+    title: metadataContent(html, "og:title") ?? metadataContent(html, "twitter:title"),
+    description: metadataContent(html, "og:description") ?? metadataContent(html, "description") ?? metadataContent(html, "twitter:description"),
+    ownerName: ownerNameFromHtml(html),
+    videoUrl: selected.url,
+    selectedQuality: selected.quality,
+    bitrate: selected.bitrate,
+    durationSec: targetDuration ?? selected.durationSec,
+    assetId: selected.assetId,
+    vencodeTag: selected.vencodeTag,
+    candidates: selectable,
+    extractedAt: (/* @__PURE__ */ new Date()).toISOString()
+  };
 }
-var import_sdk6, US_STATE_CODES, US_CITY_CENTER_ZIPS;
-var init_kernel_proxy_resolver = __esm({
-  "src/kernel-proxy-resolver.ts"() {
+async function extractFacebookOrganicVideoFromPage(page, sourceUrl, quality = "best") {
+  await page.waitForLoadState("domcontentloaded", { timeout: 15e3 }).catch(() => {
+  });
+  await page.waitForTimeout(1500);
+  const html = await page.content();
+  return extractFacebookOrganicVideo(html, page.url(), sourceUrl, quality);
+}
+var init_FacebookOrganicVideoExtractor = __esm({
+  "src/extractor/FacebookOrganicVideoExtractor.ts"() {
     "use strict";
-    import_sdk6 = __toESM(require("@onkernel/sdk"), 1);
-    init_uule();
-    US_STATE_CODES = {
-      alabama: "AL",
-      alaska: "AK",
-      arizona: "AZ",
-      arkansas: "AR",
-      california: "CA",
-      colorado: "CO",
-      connecticut: "CT",
-      delaware: "DE",
-      florida: "FL",
-      georgia: "GA",
-      hawaii: "HI",
-      idaho: "ID",
-      illinois: "IL",
-      indiana: "IN",
-      iowa: "IA",
-      kansas: "KS",
-      kentucky: "KY",
-      louisiana: "LA",
-      maine: "ME",
-      maryland: "MD",
-      massachusetts: "MA",
-      michigan: "MI",
-      minnesota: "MN",
-      mississippi: "MS",
-      missouri: "MO",
-      montana: "MT",
-      nebraska: "NE",
-      nevada: "NV",
-      "new hampshire": "NH",
-      "new jersey": "NJ",
-      "new mexico": "NM",
-      "new york": "NY",
-      "north carolina": "NC",
-      "north dakota": "ND",
-      ohio: "OH",
-      oklahoma: "OK",
-      oregon: "OR",
-      pennsylvania: "PA",
-      "rhode island": "RI",
-      "south carolina": "SC",
-      "south dakota": "SD",
-      tennessee: "TN",
-      texas: "TX",
-      utah: "UT",
-      vermont: "VT",
-      virginia: "VA",
-      washington: "WA",
-      "west virginia": "WV",
-      wisconsin: "WI",
-      wyoming: "WY"
-    };
-    US_CITY_CENTER_ZIPS = {
-      "atlanta|GA": "30303",
-      "austin|TX": "78701",
-      "baltimore|MD": "21201",
-      "boston|MA": "02108",
-      "boulder|CO": "80302",
-      "charlotte|NC": "28202",
-      "chicago|IL": "60601",
-      "colorado_springs|CO": "80903",
-      "columbus|OH": "43215",
-      "dallas|TX": "75201",
-      "denver|CO": "80202",
-      "detroit|MI": "48226",
-      "fort_collins|CO": "80524",
-      "fort_worth|TX": "76102",
-      "houston|TX": "77002",
-      "indianapolis|IN": "46204",
-      "jacksonville|FL": "32202",
-      "las_vegas|NV": "89101",
-      "los_angeles|CA": "90012",
-      "louisville|KY": "40202",
-      "loveland|CO": "80537",
-      "memphis|TN": "38103",
-      "miami|FL": "33131",
-      "minneapolis|MN": "55401",
-      "nashville|TN": "37203",
-      "new_york|NY": "10001",
-      "orlando|FL": "32801",
-      "philadelphia|PA": "19103",
-      "phoenix|AZ": "85004",
-      "portland|OR": "97205",
-      "raleigh|NC": "27601",
-      "richmond|VA": "23219",
-      "sacramento|CA": "95814",
-      "salt_lake_city|UT": "84101",
-      "san_antonio|TX": "78205",
-      "san_diego|CA": "92101",
-      "san_francisco|CA": "94103",
-      "san_jose|CA": "95113",
-      "seattle|WA": "98101"
-    };
   }
 });
 
@@ -15198,8 +15415,7 @@ async function transcribeMediaUrl(mediaUrl, markdownTitle = "# Media Transcript"
   const text = data.text ?? "";
   const chunks = data.chunks ?? [];
   const durationMs = Date.now() - startMs;
-  const audioSec = chunks.length ? chunks[chunks.length - 1]?.timestamp?.[1] ?? 0 : 0;
-  void recordVendorUsage({ vendor: "fal_wizper", model: "fal-ai/wizper", units: audioSec / 60, unitType: "audio_min" });
+  void recordVendorUsage({ vendor: "fal_wizper", model: "fal-ai/wizper", units: durationMs / 1e3, unitType: "compute_sec" });
   return {
     text,
     chunks,
@@ -16755,6 +16971,7 @@ var init_reddit_routes = __esm({
       if (!gate.ok) return c.json(concurrencyLimitExceededResponse(gate), 429, { "Retry-After": String(gate.retryAfterSeconds) });
       let debited = false;
       let refunded = false;
+      let commentDebitMc = 0;
       try {
         const { ok, balance_mc } = await debitMc(user.id, MC_COSTS.reddit_thread, LedgerOperation.REDDIT_THREAD, body.url);
         if (!ok) return c.json(insufficientBalanceResponse(balance_mc, MC_COSTS.reddit_thread), 402);
@@ -17016,6 +17233,8 @@ var init_video_routes = __esm({
     init_rates();
     init_api_auth();
     init_memory();
+    init_cost_telemetry();
+    init_cost_context();
     videoApp = new import_hono9.Hono();
     AnalyzeBodySchema = import_zod20.z.object({
       sourceUrl: import_zod20.z.string().trim().url("sourceUrl must be a direct video file URL"),
@@ -17092,6 +17311,21 @@ var init_video_routes = __esm({
           effectiveFrames: st.frameCount ?? null
         };
       }
+      if ((st.status === "done" || st.status === "failed") && typeof st.costUsd === "number" && st.costUsd > 0) {
+        const llmOp = `${LedgerOperation.VIDEO_ANALYSIS_LLM}:${runId}`;
+        if (!await ledgerExistsForOperation(user.id, llmOp)) {
+          await runWithCostContext({ ...currentCostContext(), op: "video_analysis", userId: user.id }, async () => {
+            void recordVendorUsage({ vendor: "mcp_memory_video", model: null, units: st.costUsd, unitType: "usd", error: null });
+          });
+          const surchargeMc = Math.round(st.costUsd * VIDEO_ANALYSIS_LLM_MARGIN_MULTIPLE * MC_PER_USD);
+          if (surchargeMc > 0) {
+            const { ok } = await debitMc(user.id, surchargeMc, llmOp, `real-cost-surcharge:${runId}`);
+            if (!ok) {
+              console.warn(`[video-routes] VIDEO_ANALYSIS_LLM surcharge debit failed for user ${user.id} (run ${runId}): insufficient balance for ${surchargeMc}mc`);
+            }
+          }
+        }
+      }
       return c.json({
         ok: true,
         runId,
@@ -18330,7 +18564,7 @@ var init_trustpilot_routes = __esm({
         let pagesFetched = 0;
         try {
           await driver.launch({
-            headless: true,
+            headless: false,
             kernelApiKey: browserServiceApiKey(),
             viewport: { width: 1280, height: 900 },
             locale: "en-US"
@@ -18536,7 +18770,7 @@ var init_g2_routes = __esm({
           const driver = new BrowserDriver();
           try {
             await driver.launch({
-              headless: true,
+              headless: false,
               kernelApiKey: browserServiceApiKey(),
               kernelProxyId: proxyId,
               viewport: { width: 1280, height: 900 },
@@ -19330,6 +19564,55 @@ ${headingSection}${kpoSection}${brandingSection}${bodySectionFull}${memSection}$
   }
   return { ...textResult, structuredContent };
 }
+function formatDiffPage(raw, input) {
+  const parsed = parseData(raw);
+  if ("error" in parsed) return { content: [{ type: "text", text: parsed.error }], isError: true };
+  const d = parsed.data;
+  const url = d.url ?? input.url;
+  const title = d.title ?? "Untitled";
+  const statusLine = d.status === "baseline" ? d.isReset ? "\u{1F195} **Baseline reset** \u2014 prior history discarded, this check is now the new baseline." : "\u{1F195} **Baseline captured** \u2014 first check for this URL, nothing to compare yet." : d.status === "unchanged" ? `\u2705 **No changes** since ${d.previousCheckedAt ?? "the last check"}.` : `\u{1F504} **Changed** since ${d.previousCheckedAt ?? "the last check"} \u2014 +${d.summary.linesAdded}/-${d.summary.linesRemoved} lines (${d.summary.percentChanged ?? 0}% of content).`;
+  const previewHunks = d.hunks.slice(0, DIFF_PAGE_PREVIEW_HUNKS);
+  const diffBlock = previewHunks.length ? [
+    "\n## Diff",
+    "```diff",
+    ...previewHunks.flatMap((h) => h.lines.map((l) => `${h.type === "added" ? "+" : "-"} ${l}`)),
+    "```",
+    previewHunks.length < d.hunks.length ? `*(showing ${previewHunks.length} of ${d.hunks.length} returned hunks)*` : ""
+  ].filter(Boolean).join("\n") : "";
+  const truncationNotes = [
+    d.contentTruncated ? "\n\u26A0\uFE0F Page content exceeded the storable cap and was truncated before hashing/diffing \u2014 changes past that point are not visible to this comparison." : "",
+    d.hunksTruncatedReason ? `
+\u26A0\uFE0F ${d.hunksTruncatedReason}` : ""
+  ].filter(Boolean).join("\n");
+  const tips = `
+---
+\u{1F4A1} **Tips**
+- Call \`diff_page\` again anytime to re-check \u2014 this tool is on-demand, not automatic.
+- Want the page's current content with nothing to compare? use \`extract_url\`.`;
+  const full = `# Page Diff: ${url}
+**${title}**
+
+${statusLine}${diffBlock}${truncationNotes}${tips}`;
+  return {
+    ...oneBlock(full),
+    structuredContent: {
+      url,
+      title: d.title,
+      status: d.status,
+      isReset: d.isReset,
+      previousCheckedAt: d.previousCheckedAt,
+      currentCheckedAt: d.currentCheckedAt,
+      contentHash: d.contentHash,
+      previousContentHash: d.previousContentHash,
+      summary: d.summary,
+      hunks: d.hunks,
+      contentTruncated: d.contentTruncated,
+      hunksTruncated: d.hunksTruncated,
+      hunksTruncatedReason: d.hunksTruncatedReason,
+      totalChangedLineCount: d.totalChangedLineCount
+    }
+  };
+}
 async function formatMapSiteUrls(raw, input, ctx) {
   const parsed = parseData(raw);
   if ("error" in parsed) return { content: [{ type: "text", text: parsed.error }], isError: true };
@@ -21054,7 +21337,7 @@ ${rows}` : ""
     }
   };
 }
-var import_node_fs6, import_node_os6, import_node_path8, INLINE_BUDGET_BYTES, STRUCTURED_ARRAY_CAP, reportSavingEnabled, BULK_PAGE_THRESHOLD, BULK_URL_THRESHOLD;
+var import_node_fs6, import_node_os6, import_node_path8, INLINE_BUDGET_BYTES, STRUCTURED_ARRAY_CAP, reportSavingEnabled, BULK_PAGE_THRESHOLD, BULK_URL_THRESHOLD, DIFF_PAGE_PREVIEW_HUNKS;
 var init_mcp_response_formatter = __esm({
   "src/mcp/mcp-response-formatter.ts"() {
     "use strict";
@@ -21073,6 +21356,7 @@ var init_mcp_response_formatter = __esm({
     reportSavingEnabled = true;
     BULK_PAGE_THRESHOLD = 25;
     BULK_URL_THRESHOLD = 500;
+    DIFF_PAGE_PREVIEW_HUNKS = 20;
   }
 });
 
@@ -27003,7 +27287,7 @@ var PACKAGE_VERSION;
 var init_version = __esm({
   "src/version.ts"() {
     "use strict";
-    PACKAGE_VERSION = "0.4.5";
+    PACKAGE_VERSION = "0.4.12";
   }
 });
 
@@ -27142,7 +27426,7 @@ var init_output_schema_registry = __esm({
 });
 
 // src/mcp/mcp-tool-schemas.ts
-var import_zod33, HarvestPaaInputSchema, ExtractUrlInputSchema, MapSiteUrlsInputSchema, ExtractSiteInputSchema, AuditSiteInputSchema, YoutubeHarvestInputSchema, YoutubeTranscribeInputSchema, FacebookPageIntelInputSchema, FacebookAdSearchInputSchema, RedditThreadInputSchema, VideoFrameAnalysisInputSchema, VideoFrameAnalysisStatusInputSchema, FacebookAdTranscribeInputSchema, FacebookVideoTranscribeInputSchema, GoogleAdsSearchInputSchema, GoogleAdsPageIntelInputSchema, GoogleAdsTranscribeInputSchema, InstagramProfileContentInputSchema, InstagramMediaDownloadInputSchema, MapsPlaceIntelInputSchema, TrustpilotReviewsInputSchema, G2ReviewsInputSchema, ReviewCardSchema, MapsSearchInputSchema, DirectoryWorkflowInputSchema, ArtifactPointerOutputSchema, RankTrackerModeSchema, RankTrackerBlueprintInputSchema, NullableString, MapsSearchAttemptOutput, MapsSearchOutputSchema, DirectoryMapsBusinessOutput, DirectoryWorkflowOutputSchema, RankTrackerToolPlanOutput, RankTrackerTableOutput, RankTrackerCronJobOutput, RankTrackerBlueprintOutputSchema, OrganicResultOutput, AiOverviewOutput, EntityIdsOutput, HarvestPaaOutputSchema, SearchSerpOutputSchema, ExtractUrlOutputSchema, ExtractSiteOutputSchema, AuditSiteOutputSchema, MapsPlaceIntelOutputSchema, TrustpilotReviewsOutputSchema, G2ReviewsOutputSchema, CreditsInfoOutputSchema, MapSiteUrlsOutputSchema, YoutubeHarvestOutputSchema, FacebookAdSearchOutputSchema, VideoFrameAnalysisOutputSchema, VideoFrameAnalysisStatusOutputSchema, RedditThreadOutputSchema, FacebookPageIntelOutputSchema, GoogleAdsSearchOutputSchema, GoogleAdsPageIntelOutputSchema, FacebookVideoTranscribeOutputSchema, TranscriptChunkOutput, InstagramBrowserOutput, InstagramPaginationOutput, InstagramProfileContentOutputSchema, InstagramMediaTrackOutput, InstagramDownloadOutput, InstagramMediaDownloadOutputSchema, YoutubeTranscribeOutputSchema, FacebookAdTranscribeOutputSchema, GoogleAdsTranscribeOutputSchema, CaptureSerpSnapshotOutputSchema, CaptureSerpPageSnapshotsOutputSchema, CreditsInfoInputSchema, WorkflowIdSchema2, WorkflowListInputSchema, WorkflowSuggestInputSchema, WorkflowRunInputSchema, WorkflowStepInputSchema, WorkflowStatusInputSchema, WorkflowArtifactReadInputSchema, WorkflowRecipeOutput, WorkflowDefinitionOutput, WorkflowArtifactOutput, WorkflowListOutputSchema, WorkflowSuggestOutputSchema, WorkflowRunOutputSchema, WorkflowStepOutputSchema, WorkflowStatusOutputSchema, WorkflowArtifactReadOutputSchema, SearchSerpInputSchema, CaptureSerpSnapshotInputSchema, ScreenshotInputSchema, CaptureSerpPageSnapshotsInputSchema, ReportArtifactReadInputSchema, ReportArtifactReadOutputSchema;
+var import_zod33, HarvestPaaInputSchema, ExtractUrlInputSchema, DiffPageInputSchema, MapSiteUrlsInputSchema, ExtractSiteInputSchema, AuditSiteInputSchema, YoutubeHarvestInputSchema, YoutubeTranscribeInputSchema, FacebookPageIntelInputSchema, FacebookAdSearchInputSchema, RedditThreadInputSchema, VideoFrameAnalysisInputSchema, VideoFrameAnalysisStatusInputSchema, FacebookAdTranscribeInputSchema, FacebookVideoTranscribeInputSchema, GoogleAdsSearchInputSchema, GoogleAdsPageIntelInputSchema, GoogleAdsTranscribeInputSchema, InstagramProfileContentInputSchema, InstagramMediaDownloadInputSchema, MapsPlaceIntelInputSchema, TrustpilotReviewsInputSchema, G2ReviewsInputSchema, ReviewCardSchema, MapsSearchInputSchema, DirectoryWorkflowInputSchema, ArtifactPointerOutputSchema, RankTrackerModeSchema, RankTrackerBlueprintInputSchema, NullableString, MapsSearchAttemptOutput, MapsSearchOutputSchema, DirectoryMapsBusinessOutput, DirectoryWorkflowOutputSchema, RankTrackerToolPlanOutput, RankTrackerTableOutput, RankTrackerCronJobOutput, RankTrackerBlueprintOutputSchema, OrganicResultOutput, AiOverviewOutput, EntityIdsOutput, HarvestPaaOutputSchema, SearchSerpOutputSchema, ExtractUrlOutputSchema, DiffPageOutputSchema, ExtractSiteOutputSchema, AuditSiteOutputSchema, MapsPlaceIntelOutputSchema, TrustpilotReviewsOutputSchema, G2ReviewsOutputSchema, CreditsInfoOutputSchema, MapSiteUrlsOutputSchema, YoutubeHarvestOutputSchema, FacebookAdSearchOutputSchema, VideoFrameAnalysisOutputSchema, VideoFrameAnalysisStatusOutputSchema, RedditThreadOutputSchema, FacebookPageIntelOutputSchema, GoogleAdsSearchOutputSchema, GoogleAdsPageIntelOutputSchema, FacebookVideoTranscribeOutputSchema, TranscriptChunkOutput, InstagramBrowserOutput, InstagramPaginationOutput, InstagramProfileContentOutputSchema, InstagramMediaTrackOutput, InstagramDownloadOutput, InstagramMediaDownloadOutputSchema, YoutubeTranscribeOutputSchema, FacebookAdTranscribeOutputSchema, GoogleAdsTranscribeOutputSchema, CaptureSerpSnapshotOutputSchema, CaptureSerpPageSnapshotsOutputSchema, CreditsInfoInputSchema, WorkflowIdSchema2, WorkflowListInputSchema, WorkflowSuggestInputSchema, WorkflowRunInputSchema, WorkflowStepInputSchema, WorkflowStatusInputSchema, WorkflowArtifactReadInputSchema, WorkflowRecipeOutput, WorkflowDefinitionOutput, WorkflowArtifactOutput, WorkflowListOutputSchema, WorkflowSuggestOutputSchema, WorkflowRunOutputSchema, WorkflowStepOutputSchema, WorkflowStatusOutputSchema, WorkflowArtifactReadOutputSchema, SearchSerpInputSchema, CaptureSerpSnapshotInputSchema, ScreenshotInputSchema, CaptureSerpPageSnapshotsInputSchema, ReportArtifactReadInputSchema, ReportArtifactReadOutputSchema;
 var init_mcp_tool_schemas = __esm({
   "src/mcp/mcp-tool-schemas.ts"() {
     "use strict";
@@ -27170,6 +27454,11 @@ var init_mcp_tool_schemas = __esm({
       depositToVault: import_zod33.z.boolean().default(false).describe("Save the full page content into the user's MCP Memory vault server-side, embedded for semantic recall \u2014 the full body is NOT returned to chat."),
       vaultName: import_zod33.z.string().trim().min(1).max(120).optional().describe("Optional vault to deposit into. Defaults to the user's personal vault.")
     };
+    DiffPageInputSchema = {
+      url: import_zod33.z.string().url().describe("Public http/https URL to check for changes since the last diff_page call."),
+      allowLocal: import_zod33.z.boolean().default(false).describe("Allow localhost and private-network URLs. Local development only."),
+      resetBaseline: import_zod33.z.boolean().default(false).describe("Discard any previously stored snapshot for this URL and capture the current content as a fresh baseline instead of diffing against history. Use when you deliberately want to restart change tracking.")
+    };
     MapSiteUrlsInputSchema = {
       url: import_zod33.z.string().url().describe("Public website URL or domain to crawl for internal URLs. Use before extract_site when the user asks to audit/map/crawl a site."),
       maxUrls: import_zod33.z.number().int().min(1).max(1e4).optional().describe("Maximum URLs to discover. Use 100 for normal maps, up to 10000 for a full inventory. Large maps (over 500 URLs) write the complete inventory to a local file and return only a summary plus the file path instead of the full list inline.")
@@ -27563,6 +27852,29 @@ var init_mcp_tool_schemas = __esm({
         error: import_zod33.z.string().optional()
       }).optional()
     };
+    DiffPageOutputSchema = {
+      url: import_zod33.z.string(),
+      title: NullableString,
+      status: import_zod33.z.enum(["baseline", "unchanged", "changed"]).describe('"baseline" = first-ever check for this URL, or resetBaseline was used \u2014 nothing to compare against. "unchanged" = content hash matched the stored snapshot. "changed" = a diff was computed.'),
+      isReset: import_zod33.z.boolean().describe("True only when resetBaseline discarded a real prior snapshot \u2014 distinguishes an explicit reset from a URL's true first-ever check."),
+      previousCheckedAt: NullableString.describe("ISO timestamp of the snapshot this was compared against, or null if there was none."),
+      currentCheckedAt: import_zod33.z.string().describe("ISO timestamp of this check, now stored as the new snapshot."),
+      contentHash: import_zod33.z.string().describe("sha256 of the full (untruncated) page content just captured."),
+      previousContentHash: NullableString,
+      summary: import_zod33.z.object({
+        linesAdded: import_zod33.z.number().int().min(0),
+        linesRemoved: import_zod33.z.number().int().min(0),
+        percentChanged: import_zod33.z.number().min(0).max(100).nullable().describe('Proportion of changed lines relative to the larger of the two versions. Null when status is "baseline".')
+      }),
+      hunks: import_zod33.z.array(import_zod33.z.object({
+        type: import_zod33.z.enum(["added", "removed"]).describe("Unchanged context lines are omitted \u2014 only added/removed lines are returned."),
+        lines: import_zod33.z.array(import_zod33.z.string())
+      })).describe("Ordered added/removed line hunks, capped for response size \u2014 see hunksTruncated."),
+      contentTruncated: import_zod33.z.boolean().describe("True if the scraped page exceeded the 250,000-character storable cap and was truncated before hashing/diffing/storing \u2014 changes past that point are invisible to this comparison."),
+      hunksTruncated: import_zod33.z.boolean().describe("True if the hunks list above was capped for response size \u2014 see hunksTruncatedReason."),
+      hunksTruncatedReason: NullableString,
+      totalChangedLineCount: import_zod33.z.number().int().min(0).describe("Total changed lines found before any hunksTruncated capping was applied.")
+    };
     ExtractSiteOutputSchema = {
       url: import_zod33.z.string(),
       pageCount: import_zod33.z.number().int().min(0),
@@ -28575,6 +28887,13 @@ function registerPaaExtractorMcpTools(server, executor, options = {}) {
     outputSchema: recordOutputSchema("extract_url", ExtractUrlOutputSchema),
     annotations: liveWebToolAnnotations("Single URL Extract")
   }, async (input) => formatExtractUrl(await executor.extractUrl(input), input));
+  server.registerTool("diff_page", {
+    title: "Page Change Check",
+    description: "Check whether a public URL has changed since you last checked it with this tool: scrapes the current page, diffs it against your last stored snapshot for that URL, and returns what was added or removed (or confirms no change). Stores the new snapshot as the baseline for next time \u2014 on-demand only, no automatic recurring checks. Use extract_url instead when you just want the page's current content with nothing to compare against.",
+    inputSchema: DiffPageInputSchema,
+    outputSchema: recordOutputSchema("diff_page", DiffPageOutputSchema),
+    annotations: liveWebToolAnnotations("Page Change Check")
+  }, async (input) => formatDiffPage(await executor.diffPage(input), input));
   server.registerTool("map_site_urls", {
     title: "Site URL Map",
     description: `Map/crawl a public website for a sitemap, URL inventory, or broken-link scan. Returns internal URLs with HTTP status; ${fileBehavior("maps over 500 URLs are written to a local CSV file instead of inlined.", "large results are stored as a retrievable artifact \u2014 you get an inline summary plus an artifactId for report_artifact_read.")}`,
@@ -29006,6 +29325,9 @@ var init_http_mcp_tool_executor = __esm({
       extractUrl(input) {
         return this.call("/extract-url", input);
       }
+      diffPage(input) {
+        return this.call("/diff-page", input);
+      }
       mapSiteUrls(input) {
         return this.call("/map-urls", input);
       }
@@ -37392,6 +37714,77 @@ var init_site_audit_worker = __esm({
   }
 });
 
+// src/api/page-diff.ts
+function sha256Hex(value) {
+  return (0, import_node_crypto14.createHash)("sha256").update(value).digest("hex");
+}
+function truncateForStorage(value, maxChars = MAX_SNAPSHOT_CONTENT_CHARS) {
+  if (value.length <= maxChars) return { value, truncated: false };
+  return { value: value.slice(0, maxChars), truncated: true };
+}
+function changeLines(change) {
+  const lines = change.value.split("\n");
+  if (lines.length > 0 && lines[lines.length - 1] === "") lines.pop();
+  return lines;
+}
+function diffPageContent(oldContent, newContent) {
+  const changes = (0, import_diff.diffLines)(oldContent, newContent);
+  const rawHunks = [];
+  let linesAdded = 0;
+  let linesRemoved = 0;
+  let totalChangedLineCount = 0;
+  for (const change of changes) {
+    if (!change.added && !change.removed) continue;
+    const lines = changeLines(change);
+    totalChangedLineCount += lines.length;
+    if (change.added) linesAdded += lines.length;
+    if (change.removed) linesRemoved += lines.length;
+    rawHunks.push({ type: change.added ? "added" : "removed", lines });
+  }
+  let hunksTruncated = false;
+  const hunksWithinCount = rawHunks.length > MAX_DIFF_HUNKS ? rawHunks.slice(0, MAX_DIFF_HUNKS) : rawHunks;
+  if (rawHunks.length > MAX_DIFF_HUNKS) hunksTruncated = true;
+  const hunks = [];
+  let emittedLines = 0;
+  for (const hunk of hunksWithinCount) {
+    if (emittedLines >= MAX_DIFF_LINES_PER_RESPONSE) {
+      hunksTruncated = true;
+      break;
+    }
+    const remaining = MAX_DIFF_LINES_PER_RESPONSE - emittedLines;
+    if (hunk.lines.length > remaining) {
+      hunks.push({ type: hunk.type, lines: hunk.lines.slice(0, remaining) });
+      emittedLines += remaining;
+      hunksTruncated = true;
+      break;
+    }
+    hunks.push(hunk);
+    emittedLines += hunk.lines.length;
+  }
+  const denominator = Math.max(oldContent.split("\n").length, newContent.split("\n").length, 1);
+  const percentChanged = totalChangedLineCount > 0 ? Math.min(100, Math.round(totalChangedLineCount / denominator * 1e3) / 10) : 0;
+  return {
+    hunks,
+    linesAdded,
+    linesRemoved,
+    percentChanged,
+    hunksTruncated,
+    hunksTruncatedReason: hunksTruncated ? `Showing ${emittedLines} of ${totalChangedLineCount} changed lines (capped at ${MAX_DIFF_HUNKS} hunks / ${MAX_DIFF_LINES_PER_RESPONSE} lines). Use contentHash/previousContentHash to confirm the full scope of change.` : null,
+    totalChangedLineCount
+  };
+}
+var import_node_crypto14, import_diff, MAX_SNAPSHOT_CONTENT_CHARS, MAX_DIFF_HUNKS, MAX_DIFF_LINES_PER_RESPONSE;
+var init_page_diff = __esm({
+  "src/api/page-diff.ts"() {
+    "use strict";
+    import_node_crypto14 = require("crypto");
+    import_diff = require("diff");
+    MAX_SNAPSHOT_CONTENT_CHARS = 25e4;
+    MAX_DIFF_HUNKS = 200;
+    MAX_DIFF_LINES_PER_RESPONSE = 2e3;
+  }
+});
+
 // src/api/scrape-vault-sink.ts
 async function depositScrapeToVault(user, opts) {
   try {
@@ -37826,14 +38219,17 @@ async function processJob(job) {
   running++;
   try {
     const opts = typeof job.options === "string" ? JSON.parse(job.options) : job.options;
-    const result = await harvest({
-      ...opts,
-      kernelApiKey: browserServiceApiKey(),
-      headless: true,
-      format: "json",
-      outputDir: "/tmp/paa-output-api",
-      onAttemptEvent: createHarvestAttemptRecorder(job.id, job.user_id)
-    });
+    const result = await runWithCostContext(
+      { op: opts.serpOnly ? "serp" : "paa", userId: Number(job.user_id) },
+      () => harvest({
+        ...opts,
+        kernelApiKey: browserServiceApiKey(),
+        headless: true,
+        format: "json",
+        outputDir: "/tmp/paa-output-api",
+        onAttemptEvent: createHarvestAttemptRecorder(job.id, job.user_id)
+      })
+    );
     await completeJob(job.id, result);
     const attempts = await listHarvestAttempts(job.id, job.user_id);
     if (!opts.serpOnly && typeof opts.billingHoldMc === "number") {
@@ -37899,6 +38295,7 @@ var init_worker = __esm({
     init_rates();
     init_harvest_problems();
     init_harvest_attempt_events();
+    init_cost_context();
     MAX_CONCURRENT = 2;
     running = 0;
   }
@@ -37977,9 +38374,12 @@ function opForCostPath(p) {
   }
   if (p.startsWith("/instagram/")) return "instagram";
   if (p === "/reddit/thread") return "reddit_thread";
+  if (p.startsWith("/trustpilot/")) return "trustpilot_reviews";
+  if (p.startsWith("/g2/")) return "g2_reviews";
   if (p.startsWith("/api/internal/site-architecture-auditor")) return "audit_site";
   if (p.startsWith("/api/internal/memory")) return "memory_ai";
   if (p.startsWith("/agent/")) return "browser_session";
+  if (p.startsWith("/video/")) return "video_analysis";
   return null;
 }
 function maybeProvisionInBackground(user, vaultCount) {
@@ -38089,6 +38489,7 @@ var init_server = __esm({
     import_stripe2 = __toESM(require("stripe"), 1);
     init_rates();
     init_server_schemas();
+    init_page_diff();
     init_scrape_vault_sink();
     init_scrape_blob_cleanup();
     init_schemas3();
@@ -38811,16 +39212,20 @@ var init_server = __esm({
         AbortSignal.timeout(syncTimeoutMs)
       ]);
       try {
-        const result = await harvest({
-          ...options,
-          kernelApiKey: browserServiceApiKey(),
-          headless: true,
-          format: "json",
-          outputDir: "/tmp/paa-output-api",
-          signal: syncSignal,
-          softDeadlineMs: Date.now() + Math.floor(syncTimeoutMs * 0.8),
-          onAttemptEvent: recordAttempt
-        });
+        const harvestCtx = currentCostContext();
+        const result = await runWithCostContext(
+          { ...harvestCtx, op: options.serpOnly ? "serp" : "paa", userId: user.id },
+          () => harvest({
+            ...options,
+            kernelApiKey: browserServiceApiKey(),
+            headless: true,
+            format: "json",
+            outputDir: "/tmp/paa-output-api",
+            signal: syncSignal,
+            softDeadlineMs: Date.now() + Math.floor(syncTimeoutMs * 0.8),
+            onAttemptEvent: recordAttempt
+          })
+        );
         await completeJob(jobId, result);
         const attempts = await listHarvestAttempts(jobId, user.id);
         if (!options.serpOnly) {
@@ -38988,6 +39393,91 @@ var init_server = __esm({
         await releaseConcurrencyGate(gate.lockId);
       }
     });
+    app.post("/diff-page", auth2, async (c) => {
+      const raw = await c.req.json().catch(() => ({}));
+      const bodyResult = DiffPageBodySchema.safeParse(raw);
+      if (!bodyResult.success) return c.json({ error: bodyResult.error.issues[0]?.message ?? "Invalid request" }, 400);
+      const { url, allowLocal, resetBaseline } = bodyResult.data;
+      if (!allowLocal) {
+        const checked = await validatePublicHttpUrl(url, { field: "URL" });
+        if (checked.error || !checked.parsed) return c.json({ error: checked.error ?? "Invalid URL" }, 400);
+      } else {
+        try {
+          new URL(url);
+        } catch {
+          return c.json({ error: "Invalid URL" }, 400);
+        }
+      }
+      const canonicalUrl = (() => {
+        try {
+          return new URL(url).href;
+        } catch {
+          return url;
+        }
+      })();
+      const user = c.get("user");
+      const gate = await acquireConcurrencyGate(user, "diff_page", {
+        reuseLockId: c.req.header("x-mcp-scraper-concurrency-lock"),
+        metadata: { url: canonicalUrl }
+      });
+      if (!gate.ok) return c.json(concurrencyLimitExceededResponse(gate), 429, { "Retry-After": String(gate.retryAfterSeconds) });
+      let debited = false;
+      try {
+        const { ok: dpOk, balance_mc: dpBal } = await debitMc(user.id, MC_COSTS.diff_page, LedgerOperation.DIFF_PAGE, new URL(canonicalUrl).hostname);
+        if (!dpOk) return c.json(insufficientBalanceResponse(dpBal, MC_COSTS.diff_page), 402);
+        debited = true;
+        const existingSnapshot = await getPageSnapshot(user.id, canonicalUrl);
+        const previous = resetBaseline ? null : existingSnapshot;
+        const kernelApiKey = browserServiceApiKey();
+        const result = await extractKpo({ url: canonicalUrl, kernelApiKey });
+        const currentContent = result.bodyMarkdown ?? "";
+        const currentHash = sha256Hex(currentContent);
+        const { value: storedContent, truncated: contentTruncated } = truncateForStorage(currentContent);
+        const currentCheckedAt = (/* @__PURE__ */ new Date()).toISOString();
+        await upsertPageSnapshot(user.id, canonicalUrl, {
+          contentHash: currentHash,
+          content: storedContent,
+          title: result.title ?? null,
+          contentBytes: Buffer.byteLength(currentContent, "utf8"),
+          truncated: contentTruncated
+        });
+        let status;
+        let diff = { hunks: [], linesAdded: 0, linesRemoved: 0, percentChanged: 0, hunksTruncated: false, hunksTruncatedReason: null, totalChangedLineCount: 0 };
+        if (!previous) {
+          status = "baseline";
+          diff.percentChanged = null;
+        } else if (previous.contentHash === currentHash) {
+          status = "unchanged";
+        } else {
+          status = "changed";
+          diff = diffPageContent(previous.content, storedContent);
+        }
+        await logRequestEvent({ userId: user.id, source: "diff_page", status: "done", query: canonicalUrl, result: { status } });
+        return c.json({
+          url: canonicalUrl,
+          title: result.title ?? null,
+          status,
+          isReset: !!resetBaseline && !!existingSnapshot,
+          previousCheckedAt: previous?.checkedAt ?? null,
+          currentCheckedAt,
+          contentHash: currentHash,
+          previousContentHash: previous?.contentHash ?? null,
+          summary: { linesAdded: diff.linesAdded, linesRemoved: diff.linesRemoved, percentChanged: diff.percentChanged },
+          hunks: diff.hunks,
+          contentTruncated,
+          hunksTruncated: diff.hunksTruncated,
+          hunksTruncatedReason: diff.hunksTruncatedReason,
+          totalChangedLineCount: diff.totalChangedLineCount
+        });
+      } catch (err) {
+        const msg = err instanceof Error ? err.message : String(err);
+        if (debited) await creditMc(user.id, MC_COSTS.diff_page, LedgerOperation.DIFF_PAGE, "failed call");
+        await logRequestEvent({ userId: user.id, source: "diff_page", status: "failed", query: canonicalUrl, error: msg });
+        return c.json({ error: msg }, 500);
+      } finally {
+        await releaseConcurrencyGate(gate.lockId);
+      }
+    });
     app.post("/map-urls", auth2, async (c) => {
       const raw = await c.req.json().catch(() => ({}));
       const bodyResult = MapUrlsBodySchema.safeParse(raw);