repospend 0.1.0 → 0.1.2

package/CHANGELOG.md

@@ -2,6 +2,26 @@
 
 All notable changes to RepoSpend will be documented in this file.
 
+## 0.1.2
+
+- Add bundled GPT-5.6 API-equivalent pricing for the Sol, Terra, and Luna tiers, including cache-write and cached-input rates.
+- Add bundled Claude Sonnet 5 introductory pricing through August 31, 2026, with dated model ID family pricing and a note for the standard September 1, 2026 rate.
+- Add Claude Opus 4.8 fast mode pricing and keep special fast-mode rate cards separate from regular Opus family fallback.
+- Add GitHub Copilot pricing coverage for MAI-Code-1-Flash and Kimi K2.7 Code, including suffix variant matching for local Copilot model IDs.
+- Add the OpenAI GPT-5.6 preview pricing source to the bundled pricing metadata shown in Settings.
+- Ignore local generated `marketing/` assets so release diffs stay focused on source, docs, and packaged files.
+
+## 0.1.1
+
+- Add bundled Claude Fable 5 and Claude Mythos 5 API-equivalent pricing, including dated model ID family pricing and future Fable/Mythos version fallback.
+- Price Claude cache writes from the local 5-minute vs 1-hour TTL split when transcripts expose it, with the 1-hour rate kept as the fallback for unsplit cache creation.
+- Improve the dashboard filter bar by replacing always-expanded filter groups with compact dropdowns that close when another filter opens, Escape is pressed, or the user clicks outside.
+- Move the Overview timeline and chart controls directly under the KPI row, keeping diagnostics such as Token Counting in Settings instead of the daily-glance flow.
+- Replace Settings save and local-cache actions that forced full page reloads with React refetches that preserve page context.
+- Improve chart readability with stable repo colors, positive-only minimum bar size, and a model distribution share panel for skewed model usage.
+- Reduce badge and copy noise by quieting provider badges, preserving outcome color, simplifying empty filter copy, and showing a compact filtered-view notice on pages without the filter bar.
+- Improve responsive and accessibility polish with a horizontal narrow-width nav, safer filter popover alignment, system font fallback, tabular numeric values, and reduced-motion CSS.
+
 ## 0.1.0
 
 - Add GitHub Copilot support for local OTEL exports, Copilot session-state files, and VS Code Copilot Chat transcript/debug files, leaving cost unknown when local data lacks full token splits.

package/README.md

@@ -35,11 +35,11 @@ Best for developers who want to:
 
 ## Preview
 
-![RepoSpend overview dashboard with fictional Middle-earth usage data](docs/screenshots/dashboard-overview.png)
+![Animated RepoSpend overview and sessions tour with fictional Middle-earth usage data](docs/screenshots/overview-sessions-tour.gif)
 
-Screenshots use fictional Middle-earth demo data. The Lord of the Rings themed
-repo names, sessions, prompts, token counts, and costs are intentional; no private
-repository data is shown.
+Preview media uses fictional Middle-earth demo data. The Lord of the Rings themed
+repo names, sessions, prompts, token counts, models, and costs are intentional; no
+private repository data is shown.
 
 ## What is RepoSpend?
 
@@ -196,9 +196,12 @@ RepoSpend uses normalized model-work totals:
 totalTokens = inputTokens + outputTokens + reasoningTokens
 ```
 
-Cache reads and cache writes are kept as input sub-buckets and priced once. This
-means RepoSpend token totals may look lower than tools that display cache
-reads/writes as separate addable token columns.
+Cache reads and cache writes are kept as input sub-buckets and priced once. For
+Claude, RepoSpend also uses the 5-minute vs 1-hour cache-write split recorded in
+local transcripts when available. This means RepoSpend token totals may look
+lower than tools that display cache reads/writes as separate addable token
+columns, and Claude API-equivalent cost may differ from tools that collapse all
+cache writes into one rate.
 
 For the detailed accounting model and comparison with `ccusage` and Tokscale,
 see [docs/token-accounting.md](docs/token-accounting.md).
@@ -275,6 +278,17 @@ view that is close to Claude Code's local usage files. Use RepoSpend when you
 want a local dashboard that compares AI coding usage across repos, sessions,
 models, and tools.
 
+When comparing totals, expect some intentional differences:
+
+- RepoSpend includes Claude Desktop/local-agent session files when they exist;
+  many terminal-first tools count only `~/.claude/projects`.
+- RepoSpend keeps cached input and cache writes as input sub-buckets rather than
+  adding them again to headline token totals.
+- RepoSpend prices Claude cache writes by the recorded 5-minute vs 1-hour TTL
+  split when local transcripts expose it.
+- RepoSpend separates Codex visible output from reasoning output so reasoning is
+  priced once.
+
 Generic Claude Code monitors usually focus on one source. RepoSpend is designed
 as a repo-level AI coding cost tracker: it brings together local Codex, Claude
 Code, and GitHub Copilot data, includes experimental Cursor imports when enabled,

package/dist/cli.js

@@ -9,11 +9,17 @@ function summarize(sessions) {
   const cost = sumKnownCost(sessions);
   const inputTokens = sum(sessions, "inputTokens");
   const cachedInputTokens = sum(sessions, "cachedInputTokens");
+  const cacheCreationInputTokens = optionalSum(sessions, "cacheCreationInputTokens");
+  const cacheCreationInputTokens5m = optionalSum(sessions, "cacheCreationInputTokens5m");
+  const cacheCreationInputTokens1h = optionalSum(sessions, "cacheCreationInputTokens1h");
   return {
     estimatedCostUsd: cost.knownCount > 0 ? cost.value : void 0,
     knownCostSessions: cost.knownCount,
     totalTokens: sum(sessions, "totalTokens"),
     cachedInputTokens,
+    cacheCreationInputTokens,
+    cacheCreationInputTokens5m,
+    cacheCreationInputTokens1h,
     inputTokens,
     outputTokens: sum(sessions, "outputTokens"),
     reasoningTokens: sum(sessions, "reasoningTokens"),
@@ -80,6 +86,9 @@ function groupBy(sessions, keyFn, labelFn) {
       unknownCostSessions: 0,
       inputTokens: 0,
       cachedInputTokens: 0,
+      cacheCreationInputTokens: 0,
+      cacheCreationInputTokens5m: 0,
+      cacheCreationInputTokens1h: 0,
       outputTokens: 0,
       reasoningTokens: 0,
       totalTokens: 0,
@@ -89,6 +98,9 @@ function groupBy(sessions, keyFn, labelFn) {
     };
     group.inputTokens += session.inputTokens;
     group.cachedInputTokens += session.cachedInputTokens;
+    group.cacheCreationInputTokens = (group.cacheCreationInputTokens ?? 0) + (session.cacheCreationInputTokens ?? 0);
+    group.cacheCreationInputTokens5m = (group.cacheCreationInputTokens5m ?? 0) + (session.cacheCreationInputTokens5m ?? 0);
+    group.cacheCreationInputTokens1h = (group.cacheCreationInputTokens1h ?? 0) + (session.cacheCreationInputTokens1h ?? 0);
     group.outputTokens += session.outputTokens;
     group.reasoningTokens += session.reasoningTokens;
     group.totalTokens += session.totalTokens;
@@ -128,6 +140,8 @@ function toCsv(sessions) {
     "inputTokens",
     "cachedInputTokens",
     "cacheCreationInputTokens",
+    "cacheCreationInputTokens5m",
+    "cacheCreationInputTokens1h",
     "outputTokens",
     "reasoningTokens",
     "reasoningOutputTokens",
@@ -219,11 +233,15 @@ function normalizePricingModelId(model) {
 }
 function claudePricingFamilyModel(model, isUsable = () => true) {
   const families = [
+    "claude-fable-5",
+    "claude-mythos-5",
+    "claude-opus-4-8-fast-mode",
     "claude-opus-4-7",
     "claude-opus-4-6",
     "claude-opus-4-5",
     "claude-opus-4-1",
     "claude-opus-4",
+    "claude-sonnet-5",
     "claude-sonnet-4-6",
     "claude-sonnet-4-5",
     "claude-sonnet-4",
@@ -246,11 +264,13 @@ function copilotPricingFamilyModel(model, isUsable = () => true) {
   if ((normalized === "raptor-mini" || normalized.startsWith("raptor-mini-") || normalized.startsWith("oswe-vscode")) && isUsable("raptor-mini")) return "raptor-mini";
   if ((normalized === "lark" || normalized.startsWith("lark-")) && isUsable("lark")) return "lark";
   if ((normalized === "goldeneye" || normalized.startsWith("goldeneye-")) && isUsable("goldeneye")) return "goldeneye";
+  if ((normalized === "mai-code-1-flash" || normalized.startsWith("mai-code-1-flash-")) && isUsable("mai-code-1-flash")) return "mai-code-1-flash";
+  if ((normalized === "kimi-k2.7-code" || normalized.startsWith("kimi-k2.7-code-")) && isUsable("kimi-k2.7-code")) return "kimi-k2.7-code";
   return void 0;
 }
 function parseVersionedModel(normalized) {
   if (normalized.includes("fast-mode")) return void 0;
-  const claude = normalized.match(/^(claude-(?:opus|sonnet|haiku))-(\d+)(?:-(\d+))?/);
+  const claude = normalized.match(/^(claude-(?:fable|mythos|opus|sonnet|haiku))-(\d+)(?:-(\d+))?/);
   if (claude) {
     const minor = claude[3] ? Number(claude[3]) : 0;
     return { tier: claude[1], version: Number(claude[2]) + minor / 100 };
@@ -276,7 +296,7 @@ function versionedFallbackModel(model, knownModels, isUsable = () => true) {
 }
 function isCopilotAliasModel(model) {
   const normalized = normalizePricingModelId(model);
-  return normalized.startsWith("raptor-mini") || normalized.startsWith("oswe-vscode") || normalized === "lark" || normalized.startsWith("lark-") || normalized.startsWith("goldeneye");
+  return normalized.startsWith("raptor-mini") || normalized.startsWith("oswe-vscode") || normalized === "lark" || normalized.startsWith("lark-") || normalized.startsWith("goldeneye") || normalized.startsWith("mai-code-1-flash") || normalized.startsWith("kimi-k2.7-code");
 }
 function positiveRate(value) {
   return typeof value === "number" && Number.isFinite(value) && value > 0;
@@ -303,14 +323,19 @@ var pricingInfo = {
   sourceUrl: "https://developers.openai.com/api/docs/pricing",
   sourceUrls: [
     { label: "OpenAI pricing reference", url: "https://developers.openai.com/api/docs/pricing" },
+    { label: "OpenAI GPT-5.6 preview pricing", url: "https://openai.com/index/previewing-gpt-5-6-sol/" },
     { label: "Claude pricing reference", url: "https://platform.claude.com/docs/en/about-claude/pricing" },
     { label: "GitHub Copilot model pricing reference", url: "https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing" }
   ],
   unit: "USD per 1M tokens",
-  updatedAt: "2026-05-18",
+  updatedAt: "2026-07-02",
   note: "RepoSpend estimates API-equivalent cost from local token counts and public API-style Standard pricing. This is not your actual bill; subscriptions, credits, provider terms, cache behavior, regional processing, or other billing factors can make your real cost different."
 };
 var defaultPricing = {
+  "gpt-5.6": { inputPerMillion: 5, cachedInputPerMillion: 0.5, cacheCreationInputPerMillion: 6.25, outputPerMillion: 30, reasoningOutputPerMillion: 30, note: "GPT-5.6 Sol flagship tier. OpenAI preview pricing lists Sol at $5 input / $30 output per 1M tokens, cache writes at 1.25x input, and cache reads at a 90% discount." },
+  "gpt-5.6-sol": { inputPerMillion: 5, cachedInputPerMillion: 0.5, cacheCreationInputPerMillion: 6.25, outputPerMillion: 30, reasoningOutputPerMillion: 30 },
+  "gpt-5.6-terra": { inputPerMillion: 2.5, cachedInputPerMillion: 0.25, cacheCreationInputPerMillion: 3.125, outputPerMillion: 15, reasoningOutputPerMillion: 15 },
+  "gpt-5.6-luna": { inputPerMillion: 1, cachedInputPerMillion: 0.1, cacheCreationInputPerMillion: 1.25, outputPerMillion: 6, reasoningOutputPerMillion: 6 },
   "gpt-5.5": { inputPerMillion: 5, cachedInputPerMillion: 0.5, outputPerMillion: 30, reasoningOutputPerMillion: 30 },
   "gpt-5.5-pro": { inputPerMillion: 30, outputPerMillion: 180, reasoningOutputPerMillion: 180 },
   "gpt-5.4": { inputPerMillion: 2.5, cachedInputPerMillion: 0.25, outputPerMillion: 15, reasoningOutputPerMillion: 15 },
@@ -342,18 +367,24 @@ var defaultPricing = {
   "oswe-vscode-prime": { inputPerMillion: 0.25, cachedInputPerMillion: 0.025, outputPerMillion: 2, note: "GitHub Copilot Raptor mini internal model id." },
   "lark": { inputPerMillion: 0.25, cachedInputPerMillion: 0.025, outputPerMillion: 2, note: "GitHub Copilot preview model; estimated with lightweight Copilot pricing." },
   "goldeneye": { inputPerMillion: 1.25, cachedInputPerMillion: 0.125, outputPerMillion: 10, note: "GitHub Copilot fine-tuned model using GPT-5.1-Codex pricing." },
-  "claude-opus-4-8": { inputPerMillion: 5, cacheCreationInputPerMillion: 6.25, cachedInputPerMillion: 0.5, outputPerMillion: 25 },
-  "claude-opus-4-7": { inputPerMillion: 5, cacheCreationInputPerMillion: 6.25, cachedInputPerMillion: 0.5, outputPerMillion: 25 },
-  "claude-opus-4-6": { inputPerMillion: 5, cacheCreationInputPerMillion: 6.25, cachedInputPerMillion: 0.5, outputPerMillion: 25 },
-  "claude-opus-4-6-fast-mode": { inputPerMillion: 0, cacheCreationInputPerMillion: 0, cachedInputPerMillion: 0, outputPerMillion: 0, note: "GitHub Copilot supported model, but public per-token pricing is not listed separately. RepoSpend leaves API-equivalent pricing unset." },
-  "claude-opus-4-5": { inputPerMillion: 5, cacheCreationInputPerMillion: 6.25, cachedInputPerMillion: 0.5, outputPerMillion: 25 },
-  "claude-opus-4-1": { inputPerMillion: 15, cacheCreationInputPerMillion: 18.75, cachedInputPerMillion: 1.5, outputPerMillion: 75 },
-  "claude-opus-4": { inputPerMillion: 15, cacheCreationInputPerMillion: 18.75, cachedInputPerMillion: 1.5, outputPerMillion: 75 },
-  "claude-sonnet-4-6": { inputPerMillion: 3, cacheCreationInputPerMillion: 3.75, cachedInputPerMillion: 0.3, outputPerMillion: 15 },
-  "claude-sonnet-4-5": { inputPerMillion: 3, cacheCreationInputPerMillion: 3.75, cachedInputPerMillion: 0.3, outputPerMillion: 15 },
-  "claude-sonnet-4": { inputPerMillion: 3, cacheCreationInputPerMillion: 3.75, cachedInputPerMillion: 0.3, outputPerMillion: 15 },
-  "claude-haiku-4-5": { inputPerMillion: 1, cacheCreationInputPerMillion: 1.25, cachedInputPerMillion: 0.1, outputPerMillion: 5 },
-  "claude-3-5-haiku": { inputPerMillion: 0.8, cacheCreationInputPerMillion: 1, cachedInputPerMillion: 0.08, outputPerMillion: 4 }
+  "mai-code-1-flash": { inputPerMillion: 0.75, cachedInputPerMillion: 0.075, outputPerMillion: 4.5, note: "GitHub Copilot supported Microsoft model; API-equivalent estimate, not a Copilot bill." },
+  "kimi-k2.7-code": { inputPerMillion: 0.95, cachedInputPerMillion: 0.19, outputPerMillion: 4, note: "GitHub Copilot supported Moonshot AI model; API-equivalent estimate, not a Copilot bill." },
+  "claude-fable-5": { inputPerMillion: 10, cacheCreationInput5mPerMillion: 12.5, cacheCreationInput1hPerMillion: 20, cacheCreationInputPerMillion: 20, cachedInputPerMillion: 1, outputPerMillion: 50 },
+  "claude-mythos-5": { inputPerMillion: 10, cacheCreationInput5mPerMillion: 12.5, cacheCreationInput1hPerMillion: 20, cacheCreationInputPerMillion: 20, cachedInputPerMillion: 1, outputPerMillion: 50, note: "Limited availability Anthropic model; same public API-equivalent pricing as Claude Fable 5." },
+  "claude-opus-4-8": { inputPerMillion: 5, cacheCreationInput5mPerMillion: 6.25, cacheCreationInput1hPerMillion: 10, cacheCreationInputPerMillion: 10, cachedInputPerMillion: 0.5, outputPerMillion: 25 },
+  "claude-opus-4-8-fast-mode": { inputPerMillion: 10, cacheCreationInput5mPerMillion: 12.5, cacheCreationInput1hPerMillion: 20, cacheCreationInputPerMillion: 12.5, cachedInputPerMillion: 1, outputPerMillion: 50, note: "Claude Opus 4.8 fast mode research preview pricing. Prompt caching multipliers apply on top of fast mode pricing." },
+  "claude-opus-4-7": { inputPerMillion: 5, cacheCreationInput5mPerMillion: 6.25, cacheCreationInput1hPerMillion: 10, cacheCreationInputPerMillion: 10, cachedInputPerMillion: 0.5, outputPerMillion: 25 },
+  "claude-opus-4-6": { inputPerMillion: 5, cacheCreationInput5mPerMillion: 6.25, cacheCreationInput1hPerMillion: 10, cacheCreationInputPerMillion: 10, cachedInputPerMillion: 0.5, outputPerMillion: 25 },
+  "claude-opus-4-6-fast-mode": { inputPerMillion: 0, cacheCreationInput5mPerMillion: 0, cacheCreationInput1hPerMillion: 0, cacheCreationInputPerMillion: 0, cachedInputPerMillion: 0, outputPerMillion: 0, note: "GitHub Copilot supported model, but public per-token pricing is not listed separately. RepoSpend leaves API-equivalent pricing unset." },
+  "claude-opus-4-5": { inputPerMillion: 5, cacheCreationInput5mPerMillion: 6.25, cacheCreationInput1hPerMillion: 10, cacheCreationInputPerMillion: 10, cachedInputPerMillion: 0.5, outputPerMillion: 25 },
+  "claude-opus-4-1": { inputPerMillion: 15, cacheCreationInput5mPerMillion: 18.75, cacheCreationInput1hPerMillion: 30, cacheCreationInputPerMillion: 30, cachedInputPerMillion: 1.5, outputPerMillion: 75 },
+  "claude-opus-4": { inputPerMillion: 15, cacheCreationInput5mPerMillion: 18.75, cacheCreationInput1hPerMillion: 30, cacheCreationInputPerMillion: 30, cachedInputPerMillion: 1.5, outputPerMillion: 75 },
+  "claude-sonnet-5": { inputPerMillion: 2, cacheCreationInput5mPerMillion: 2.5, cacheCreationInput1hPerMillion: 4, cacheCreationInputPerMillion: 4, cachedInputPerMillion: 0.2, outputPerMillion: 10, note: "Anthropic introductory pricing through August 31, 2026. Standard pricing from September 1, 2026 is $3 input / $15 output per 1M tokens." },
+  "claude-sonnet-4-6": { inputPerMillion: 3, cacheCreationInput5mPerMillion: 3.75, cacheCreationInput1hPerMillion: 6, cacheCreationInputPerMillion: 6, cachedInputPerMillion: 0.3, outputPerMillion: 15 },
+  "claude-sonnet-4-5": { inputPerMillion: 3, cacheCreationInput5mPerMillion: 3.75, cacheCreationInput1hPerMillion: 6, cacheCreationInputPerMillion: 6, cachedInputPerMillion: 0.3, outputPerMillion: 15 },
+  "claude-sonnet-4": { inputPerMillion: 3, cacheCreationInput5mPerMillion: 3.75, cacheCreationInput1hPerMillion: 6, cacheCreationInputPerMillion: 6, cachedInputPerMillion: 0.3, outputPerMillion: 15 },
+  "claude-haiku-4-5": { inputPerMillion: 1, cacheCreationInput5mPerMillion: 1.25, cacheCreationInput1hPerMillion: 2, cacheCreationInputPerMillion: 2, cachedInputPerMillion: 0.1, outputPerMillion: 5 },
+  "claude-3-5-haiku": { inputPerMillion: 0.8, cacheCreationInput5mPerMillion: 1, cacheCreationInput1hPerMillion: 1.6, cacheCreationInputPerMillion: 1.6, cachedInputPerMillion: 0.08, outputPerMillion: 4 }
 };
 function loadPricingTable(pricingPath) {
   if (!pricingPath) {
@@ -385,14 +416,23 @@ function calculateCostUsd(usage, pricing) {
   if (!modelPricing) {
     return void 0;
   }
-  const cacheCreationInput = usage.cacheCreationInputTokens ?? 0;
+  const cacheCreationInput5m = usage.cacheCreationInputTokens5m ?? 0;
+  const cacheCreationInput1h = usage.cacheCreationInputTokens1h ?? 0;
+  const splitCacheCreationInput = cacheCreationInput5m + cacheCreationInput1h;
+  const cacheCreationInput = usage.cacheCreationInputTokens ?? splitCacheCreationInput;
+  const unclassifiedCacheCreationInput = Math.max(cacheCreationInput - splitCacheCreationInput, 0);
   const billableInput = Math.max(usage.inputTokens - usage.cachedInputTokens - cacheCreationInput, 0);
   const cachedRate = modelPricing.cachedInputPerMillion ?? modelPricing.inputPerMillion;
-  const cacheCreationRate = modelPricing.cacheCreationInputPerMillion ?? modelPricing.inputPerMillion;
+  const cacheCreationRate = firstPositiveRate(modelPricing.cacheCreationInputPerMillion, modelPricing.cacheCreationInput1hPerMillion, modelPricing.cacheCreationInput5mPerMillion, modelPricing.inputPerMillion);
+  const cacheCreation5mRate = firstPositiveRate(modelPricing.cacheCreationInput5mPerMillion, modelPricing.cacheCreationInputPerMillion, modelPricing.inputPerMillion);
+  const cacheCreation1hRate = firstPositiveRate(modelPricing.cacheCreationInput1hPerMillion, modelPricing.cacheCreationInputPerMillion, modelPricing.inputPerMillion);
   const reasoningRate = modelPricing.reasoningOutputPerMillion ?? modelPricing.outputPerMillion;
-  const cost = billableInput / 1e6 * modelPricing.inputPerMillion + usage.cachedInputTokens / 1e6 * cachedRate + cacheCreationInput / 1e6 * cacheCreationRate + usage.outputTokens / 1e6 * modelPricing.outputPerMillion + usage.reasoningTokens / 1e6 * reasoningRate;
+  const cost = billableInput / 1e6 * modelPricing.inputPerMillion + usage.cachedInputTokens / 1e6 * cachedRate + cacheCreationInput5m / 1e6 * cacheCreation5mRate + cacheCreationInput1h / 1e6 * cacheCreation1hRate + unclassifiedCacheCreationInput / 1e6 * cacheCreationRate + usage.outputTokens / 1e6 * modelPricing.outputPerMillion + usage.reasoningTokens / 1e6 * reasoningRate;
   return Number(cost.toFixed(6));
 }
+function firstPositiveRate(...rates) {
+  return rates.find((rate) => typeof rate === "number" && Number.isFinite(rate) && rate > 0) ?? 0;
+}
 
 // packages/core/src/cache.ts
 var activeCacheVersion = "parse-v1";
@@ -2795,6 +2835,8 @@ function claudeBreakdownToUsage(file, index, options, breakdown, idSuffix = "")
     inputTokens: breakdown.inputTokens,
     cachedInputTokens: breakdown.cachedInputTokens,
     cacheCreationInputTokens: breakdown.cacheCreationInputTokens,
+    cacheCreationInputTokens5m: breakdown.cacheCreationInputTokens5m,
+    cacheCreationInputTokens1h: breakdown.cacheCreationInputTokens1h,
     outputTokens: breakdown.outputTokens,
     reasoningTokens: 0,
     reasoningOutputTokens: 0,
@@ -2817,12 +2859,14 @@ function claudeBreakdownToUsage(file, index, options, breakdown, idSuffix = "")
     serviceTierSource: serviceTierSummary.serviceTier ? "session_usage" : void 0,
     serviceTierConfidence: serviceTierSummary.serviceTier ? "high" : void 0,
     serviceTierDetail: serviceTierSummary.serviceTierDetail,
-    sourceMetadata: breakdown.serviceTiers.length || breakdown.serviceSpeeds.length ? {
+    sourceMetadata: breakdown.serviceTiers.length || breakdown.serviceSpeeds.length || breakdown.cacheCreationInputTokens5m > 0 || breakdown.cacheCreationInputTokens1h > 0 ? {
       claude: {
         serviceTiers: breakdown.serviceTiers,
         speeds: breakdown.serviceSpeeds,
         serviceTier: serviceTierSummary.serviceTier,
-        speed: serviceSpeedSummary.serviceTier
+        speed: serviceSpeedSummary.serviceTier,
+        cacheCreationInputTokens5m: breakdown.cacheCreationInputTokens5m,
+        cacheCreationInputTokens1h: breakdown.cacheCreationInputTokens1h
       }
     } : void 0,
     userPromptCount: breakdown.userPromptCount,
@@ -2951,19 +2995,30 @@ function occurrenceKey(filePath, recordIndex) {
   return `${path6.resolve(filePath)}:${recordIndex}`;
 }
 function cloneUsage(usage) {
+  const cacheCreation = firstObject4(usage.cache_creation);
   return {
     input_tokens: numberValue(usage.input_tokens),
     cache_read_input_tokens: numberValue(usage.cache_read_input_tokens),
     cache_creation_input_tokens: numberValue(usage.cache_creation_input_tokens),
+    cache_creation: cacheCreation ? {
+      ephemeral_5m_input_tokens: numberValue(cacheCreation.ephemeral_5m_input_tokens),
+      ephemeral_1h_input_tokens: numberValue(cacheCreation.ephemeral_1h_input_tokens)
+    } : void 0,
     output_tokens: numberValue(usage.output_tokens),
     service_tier: stringValue3(usage.service_tier),
     speed: stringValue3(usage.speed)
   };
 }
 function mergeUsageMax(target, usage) {
+  const targetCacheCreation = firstObject4(target.cache_creation) ?? {};
+  const usageCacheCreation = firstObject4(usage.cache_creation) ?? {};
   target.input_tokens = Math.max(numberValue(target.input_tokens), numberValue(usage.input_tokens));
   target.cache_read_input_tokens = Math.max(numberValue(target.cache_read_input_tokens), numberValue(usage.cache_read_input_tokens));
   target.cache_creation_input_tokens = Math.max(numberValue(target.cache_creation_input_tokens), numberValue(usage.cache_creation_input_tokens));
+  target.cache_creation = {
+    ephemeral_5m_input_tokens: Math.max(numberValue(targetCacheCreation.ephemeral_5m_input_tokens), numberValue(usageCacheCreation.ephemeral_5m_input_tokens)),
+    ephemeral_1h_input_tokens: Math.max(numberValue(targetCacheCreation.ephemeral_1h_input_tokens), numberValue(usageCacheCreation.ephemeral_1h_input_tokens))
+  };
   target.output_tokens = Math.max(numberValue(target.output_tokens), numberValue(usage.output_tokens));
   target.service_tier ??= stringValue3(usage.service_tier);
   target.speed ??= stringValue3(usage.speed);
@@ -2971,11 +3026,17 @@ function mergeUsageMax(target, usage) {
 function applyUsage(breakdown, usage) {
   const input = numberValue(usage.input_tokens);
   const cacheRead = numberValue(usage.cache_read_input_tokens);
-  const cacheCreation = numberValue(usage.cache_creation_input_tokens);
+  const cacheCreationBreakdown = firstObject4(usage.cache_creation);
+  const cacheCreation5m = numberValue(cacheCreationBreakdown?.ephemeral_5m_input_tokens);
+  const cacheCreation1h = numberValue(cacheCreationBreakdown?.ephemeral_1h_input_tokens);
+  const splitCacheCreation = cacheCreation5m + cacheCreation1h;
+  const cacheCreation = Math.max(numberValue(usage.cache_creation_input_tokens), splitCacheCreation);
   const output = numberValue(usage.output_tokens);
   breakdown.inputTokens += input + cacheRead + cacheCreation;
   breakdown.cachedInputTokens += cacheRead;
   breakdown.cacheCreationInputTokens += cacheCreation;
+  breakdown.cacheCreationInputTokens5m += cacheCreation5m;
+  breakdown.cacheCreationInputTokens1h += cacheCreation1h;
   breakdown.outputTokens += output;
   breakdown.usageCount += 1;
   pushUnique(breakdown.serviceTiers, normalizedServiceTier(usage.service_tier));
@@ -3030,7 +3091,8 @@ function dateKey2(timestamp3) {
   return timestamp3?.slice(0, 10) ?? "unknown-date";
 }
 function usageTotal(usage) {
-  return numberValue(usage.input_tokens) + numberValue(usage.cache_read_input_tokens) + numberValue(usage.cache_creation_input_tokens) + numberValue(usage.output_tokens);
+  const cacheCreation = firstObject4(usage.cache_creation);
+  return numberValue(usage.input_tokens) + numberValue(usage.cache_read_input_tokens) + Math.max(numberValue(usage.cache_creation_input_tokens), numberValue(cacheCreation?.ephemeral_5m_input_tokens) + numberValue(cacheCreation?.ephemeral_1h_input_tokens)) + numberValue(usage.output_tokens);
 }
 function applyContentSignals(breakdown, content) {
   if (!Array.isArray(content)) return;
@@ -3073,6 +3135,8 @@ function emptyBreakdown2() {
     inputTokens: 0,
     cachedInputTokens: 0,
     cacheCreationInputTokens: 0,
+    cacheCreationInputTokens5m: 0,
+    cacheCreationInputTokens1h: 0,
     outputTokens: 0,
     totalTokens: 0,
     usageCount: 0,
@@ -5212,12 +5276,7 @@ import Fastify from "fastify";
 
 // apps/server/src/demo/fixtures.ts
 var baseDate = "2026-05-18";
-var demoPricing = {
-  ...defaultPricing,
-  "gpt-5.3-codex": { inputPerMillion: 1.75, cachedInputPerMillion: 0.175, outputPerMillion: 14, reasoningOutputPerMillion: 14 },
-  "gpt-5.3-codex-spark": { inputPerMillion: 1.75, cachedInputPerMillion: 0.175, outputPerMillion: 14, reasoningOutputPerMillion: 14 },
-  "claude-haiku-4-5-20251001": { inputPerMillion: 1, cacheCreationInputPerMillion: 1.25, cachedInputPerMillion: 0.1, outputPerMillion: 5 }
-};
+var demoPricing = defaultPricing;
 function issue(command2, category, severity, impact, reason) {
   return {
     command: command2,
@@ -5375,11 +5434,11 @@ var demoInputs = [
   {
     id: "lotr-session-003",
     repo: "shire-mobile",
-    title: "Investigate token spike after second breakfast",
+    title: "Investigate Opus 4.8 token spike after second breakfast",
     sourceClient: "claude",
     sourceApp: "Claude Code",
     sourceAppRaw: "claude-code",
-    model: "claude-opus-4-7",
+    model: "claude-opus-4-8",
     provider: "anthropic",
     hour: 11,
     minute: 5,
@@ -5458,11 +5517,11 @@ var demoInputs = [
   {
     id: "lotr-session-006",
     repo: "rivendell-dashboard",
-    title: "Refactor the council-of-elrond planner",
+    title: "Refactor the council-of-elrond planner with Fable 5",
     sourceClient: "claude",
     sourceApp: "Claude Desktop App",
     sourceAppRaw: "claude-desktop-local-agent",
-    model: "claude-sonnet-4-6",
+    model: "claude-fable-5",
     provider: "anthropic",
     hour: 14,
     minute: 15,
@@ -5594,11 +5653,11 @@ var demoInputs = [
   {
     id: "lotr-session-009",
     repo: "minas-tirith-admin",
-    title: "Tighten city-gate roles before the beacons launch",
+    title: "Tighten city-gate roles with Mythos 5 before the beacons launch",
     sourceClient: "claude",
     sourceApp: "Claude Code",
     sourceAppRaw: "claude-code",
-    model: "claude-haiku-4-5-20251001",
+    model: "claude-mythos-5",
     provider: "anthropic",
     hour: 17,
     minute: 5,
@@ -5750,11 +5809,11 @@ var demoInputs = [
   {
     id: "lotr-session-014",
     repo: "shire-mobile",
-    title: "Make pantry sync work offline in Bag End",
+    title: "Make pantry sync work offline in Bag End with Sonnet 4.6",
     sourceClient: "claude",
     sourceApp: "Claude Code",
     sourceAppRaw: "claude-code",
-    model: "claude-haiku-4-5-20251001",
+    model: "claude-sonnet-4-6-20260601",
     provider: "anthropic",
     hour: 20,
     minute: 45,
@@ -6297,9 +6356,13 @@ function validatePricingBody(body) {
       outputPerMillion
     };
     const cachedInputPerMillion = finiteNumber(row.cachedInputPerMillion);
+    const cacheCreationInput5mPerMillion = finiteNumber(row.cacheCreationInput5mPerMillion);
+    const cacheCreationInput1hPerMillion = finiteNumber(row.cacheCreationInput1hPerMillion);
     const cacheCreationInputPerMillion = finiteNumber(row.cacheCreationInputPerMillion);
     const reasoningOutputPerMillion = finiteNumber(row.reasoningOutputPerMillion);
     if (cachedInputPerMillion !== void 0) modelPricing.cachedInputPerMillion = cachedInputPerMillion;
+    if (cacheCreationInput5mPerMillion !== void 0) modelPricing.cacheCreationInput5mPerMillion = cacheCreationInput5mPerMillion;
+    if (cacheCreationInput1hPerMillion !== void 0) modelPricing.cacheCreationInput1hPerMillion = cacheCreationInput1hPerMillion;
     if (cacheCreationInputPerMillion !== void 0) modelPricing.cacheCreationInputPerMillion = cacheCreationInputPerMillion;
     if (reasoningOutputPerMillion !== void 0) modelPricing.reasoningOutputPerMillion = reasoningOutputPerMillion;
     if (typeof row.note === "string") modelPricing.note = row.note;

package/docs/data-sources.md

@@ -30,6 +30,13 @@ RepoSpend scans:
 
 Claude Code session transcripts are parsed from JSONL files. Useful fields include `sessionId`, `cwd`, `gitBranch`, `timestamp`, `entrypoint`, `message.model`, and `message.usage`. When `message.usage` is present, RepoSpend sums Claude assistant usage records directly. When token counts or model names are absent in a transcript, sessions remain visible with unknown cost.
 
+RepoSpend intentionally includes Claude Desktop/local-agent session roots when
+they exist. Tools such as `ccusage` and Tokscale commonly focus on
+`~/.claude/projects`, so RepoSpend's Claude totals can be higher when local-agent
+sessions are present. In one all-time local audit, RepoSpend's
+`~/.claude/projects` slice matched `ccusage`, while the difference came from
+`~/Library/Application Support/Claude/local-agent-mode-sessions`.
+
 RepoSpend checks `~/.claude/history.jsonl` only for source status/counting. History-only entries are not imported into usage analytics because they do not include reliable token, model, or transcript data.
 
 ## GitHub Copilot

package/docs/llms.txt

@@ -53,6 +53,7 @@ Positioning:
 - RepoSpend complements CLI tools such as ccusage.
 - ccusage is useful for terminal-first Claude Code totals and daily breakdowns.
 - RepoSpend is useful for visual repo-level analytics, session inspection, source/tool comparison, model mix, token shape, cache reuse, exports, and command/agent friction.
+- RepoSpend totals can intentionally differ from ccusage or Tokscale because it includes Claude Desktop/local-agent session roots when present, treats cache reads/writes as input sub-buckets, prices Claude cache writes by recorded 5-minute vs 1-hour TTL, and separates Codex visible output from reasoning output.
 
 Cost language:
 - RepoSpend shows estimated API-equivalent cost.

package/docs/pricing.md

@@ -4,10 +4,24 @@ RepoSpend estimates API-equivalent cost from a local pricing table. The bundled
 
 The bundled table is seeded from public OpenAI, Anthropic, Google, and GitHub Copilot model references and is expressed as USD per 1M tokens. Pricing changes over time, so treat RepoSpend costs as API-equivalent estimates rather than invoice-grade accounting.
 
+The bundled defaults use currently published standard API prices unless a provider has an active introductory API rate. For example, Claude Sonnet 5 uses Anthropic's $2 input and $10 output introductory rate through August 31, 2026, with a pricing-table note for the $3 input and $15 output standard rate that starts September 1, 2026.
+
+Some GitHub Copilot model rows are hosted or fine-tuned vendor models such as Raptor mini, MAI-Code-1-Flash, and Kimi K2.7 Code. RepoSpend prices those rows from GitHub's public per-token table as API-equivalent estimates, not as a Copilot invoice.
+
+For Claude models, RepoSpend uses Anthropic's cache-write TTL split when local
+Claude Code usage reports expose it. Tokens recorded under
+`cache_creation.ephemeral_5m_input_tokens` use the 5-minute cache-write rate, and
+tokens recorded under `cache_creation.ephemeral_1h_input_tokens` use the 1-hour
+cache-write rate. If a source only exposes the older aggregate
+`cache_creation_input_tokens` field, RepoSpend falls back to the model's generic
+cache-write rate.
+
 Each model can define:
 
 - input tokens
-- cache write input tokens
+- 5-minute cache write input tokens
+- 1-hour cache write input tokens
+- generic cache write input tokens for sources without a TTL split
 - cached input tokens
 - output tokens
 - reasoning output tokens
@@ -25,8 +39,20 @@ from visible output before storing and pricing both buckets. If a comparison too
 shows output inclusive of reasoning and also prices reasoning separately, its
 cost will be higher because reasoning is counted twice.
 
+This is a known reason RepoSpend Codex API-equivalent cost can be lower than a
+tool whose output bucket remains inclusive of reasoning while also exposing a
+separate reasoning bucket. Compare visible output and reasoning separately before
+treating a cost delta as a pricing-table problem.
+
 See [token-accounting.md](token-accounting.md) for the detailed comparison model.
 
+This can differ from tools or older `ccusage` versions that price all Claude
+cache creation tokens with one cache-write rate. A single-rate calculation can
+understate sessions that mostly used Anthropic's 1-hour cache writes or overstate
+sessions that mostly used 5-minute cache writes. RepoSpend prices the buckets
+recorded in the local Claude transcript instead of forcing every cache write into
+one column.
+
 When an exact model id is not present in the pricing table, RepoSpend first tries conservative family matching for known provider naming patterns. For example, a nearby newer Claude Opus 4.x or GPT 5.x variant can inherit the closest older bundled rate so the dashboard stays useful while public rate cards catch up. Inherited rates are labeled in Settings.
 
 If RepoSpend cannot resolve a usable rate, it still displays token totals and marks cost as unknown. Unknown pricing does not stop scans, dashboard responses, CLI output, or exports.

package/docs/token-accounting.md

@@ -5,6 +5,7 @@ RepoSpend uses one normalized token shape across local clients:
 - `inputTokens` is total input for the request/session. When a source reports cache reads or cache writes separately, RepoSpend includes them in `inputTokens` and also stores them in sub-buckets.
 - `cachedInputTokens` is the cache-read portion of input.
 - `cacheCreationInputTokens` is the cache-write portion of input, when the source exposes it.
+- `cacheCreationInputTokens5m` and `cacheCreationInputTokens1h` preserve Anthropic's Claude cache-write TTL split when local transcripts expose `usage.cache_creation`.
 - `outputTokens` is visible/non-reasoning output.
 - `reasoningTokens` is reasoning output, when the source exposes it.
 - `totalTokens = inputTokens + outputTokens + reasoningTokens`.
@@ -37,6 +38,39 @@ cache write input = cacheCreationInputTokens
 
 Adding cached tokens again would inflate token totals and double-charge cached input in API-equivalent cost.
 
+## Claude Source Scope
+
+RepoSpend scans both Claude project transcripts and Claude Desktop/local-agent
+session roots:
+
+```text
+~/.claude/projects
+~/.config/claude/projects
+~/Library/Application Support/Claude/local-agent-mode-sessions
+~/.config/Claude/local-agent-mode-sessions
+```
+
+Many terminal-first usage tools focus on `~/.claude/projects`. If local-agent
+session files exist, RepoSpend can show higher Claude totals than `ccusage` or
+Tokscale without a parser bug. When comparing tools, first check whether the
+same Claude roots are included.
+
+For Claude, RepoSpend also preserves the cache-write TTL split when it is present
+in local logs:
+
+```text
+5-minute cache write = cacheCreationInputTokens5m
+1-hour cache write = cacheCreationInputTokens1h
+unclassified cache write = cacheCreationInputTokens - cacheCreationInputTokens5m - cacheCreationInputTokens1h
+```
+
+This is one reason RepoSpend API-equivalent cost can differ from `ccusage`.
+Anthropic prices 5-minute cache writes at a different rate than 1-hour cache
+writes. RepoSpend applies the recorded TTL-specific rate for each bucket; tools
+that collapse all Claude cache creation tokens into a single cache-write column
+can drift when a session mixes TTLs or when the chosen single rate does not match
+the session's actual cache usage.
+
 ## Reasoning Tokens
 
 RepoSpend keeps reasoning tokens separate when the local source exposes them. The accurate treatment depends on the source's raw usage shape:

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "repospend",
-  "version": "0.1.0",
+  "version": "0.1.2",
   "description": "Local-first dashboard for tracking AI coding token usage and API-equivalent spend by repository, session, model, and tool.",
   "license": "Apache-2.0",
   "author": "Mehmet Mustafa Demir",
@@ -45,7 +45,9 @@
     "typecheck": "pnpm --filter @repospend/types --filter @repospend/core build && pnpm --stream --filter @repospend/types --filter @repospend/core --filter @repospend/server --filter @repospend/web typecheck",
     "clean": "pnpm -r clean && rm -rf dist web-dist",
     "rebuild:native": "pnpm rebuild better-sqlite3",
-    "doctor:native": "node scripts/check-native.mjs"
+    "doctor:native": "node scripts/check-native.mjs",
+    "media:capture": "node scripts/capture-release-media.mjs",
+    "media:producthunt": "node scripts/generate-producthunt-media.mjs"
   },
   "keywords": [
     "codex",
@@ -81,6 +83,9 @@
     "@types/node": "^22.19.1",
     "esbuild": "^0.28.0",
     "eslint": "^9.39.1",
+    "gifenc": "^1.0.3",
+    "playwright": "^1.60.0",
+    "sharp": "^0.35.1",
     "typescript": "^5.9.3",
     "typescript-eslint": "^8.46.4",
     "vitest": "^3.2.4"