slash-tokens 1.1.2 → 1.2.0

package/README.md

@@ -14,12 +14,30 @@ const check = preflight(prompt, 'claude-opus')
 // tokens: 47,000 | cost: $0.71 | 11 cheaper options | save 99%
 ```
 
-Or one line — every API call checked pre-call:
+Or one line — every API call optimized pre-call:
 
 ```js
 import 'slash-tokens/auto'
 ```
 
+Every call routed to the cheapest model that can execute it efficiently. Session summary on exit:
+
+```
+[slash] Session: 47 calls | 23 routed | $0.89 salvaged — The more you build, the more you save
+```
+
+## Next.js Starter
+
+One-click deploy with Vercel AI SDK + slash-tokens built in:
+
+[![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https://github.com/Wolfe-Jam/slash-nextjs&project-name=slash-nextjs&env=ANTHROPIC_API_KEY)
+
+→ [slash-nextjs on GitHub](https://github.com/Wolfe-Jam/slash-nextjs)
+
+## Dashboard
+
+Track savings across all your apps. Free key at [mcpaas.live/slash/setup](https://mcpaas.live/slash/setup)
+
 Full docs, examples, and model pricing at **[GitHub](https://github.com/Wolfe-Jam/slash-tokens)**
 
 [slashtokens.com](https://slashtokens.com) | MIT

package/dist/auto.js

@@ -19,6 +19,20 @@ const R = '\x1b[0m';
 const B = '\x1b[1m';
 const GREEN = '\x1b[38;2;74;222;128m';
 const GOLD = '\x1b[38;2;230;161;65m';
+const DIM = '\x1b[2m';
+// Session summary on exit — print total saved when the process ends
+function printSummary() {
+    if (totalCalls === 0)
+        return;
+    const savedStr = totalSalvaged > 0 ? `${GREEN}${B}$${totalSalvaged.toFixed(4)} salvaged${R}` : `$0 salvaged`;
+    const routedStr = totalRouted > 0 ? ` | ${GOLD}${totalRouted} routed${R}` : '';
+    console.log(`\n${GREEN}[slash]${R} Session: ${totalCalls} calls${routedStr} | ${savedStr} ${DIM}— The more you build, the more you save${R}`);
+}
+if (typeof process !== 'undefined') {
+    process.once('exit', printSummary);
+    process.once('SIGINT', () => { printSummary(); process.exit(0); });
+    process.once('SIGTERM', () => { printSummary(); process.exit(0); });
+}
 patchFetch();
 onIntercept((event) => {
     totalCalls++;

package/dist/config.js

@@ -24,7 +24,7 @@ export function resolveKey(perCallKey) {
     return key;
 }
 export function getEndpoint() {
-    return _endpoint;
+    return process.env.SLASH_ENDPOINT || _endpoint;
 }
 export function hasKey() {
     return !!(_key || process.env.SLASH_KEY);

package/dist/intercept.js

@@ -37,7 +37,11 @@ const AI_ENDPOINTS = [
     {
         pattern: /generativelanguage\.googleapis\.com/,
         provider: 'Google',
-        modelExtractor: (body) => 'gemini-2.5-flash',
+        modelExtractor: (_body, url) => {
+            // Model is in the URL path: /v1beta/models/gemini-2.0-flash:generateContent
+            const match = url?.match(/\/models\/([^/:]+)/);
+            return match ? match[1] : 'gemini-2.5-flash';
+        },
     },
     {
         pattern: /api\.x\.ai/,
@@ -140,7 +144,7 @@ export function patchFetch() {
                 if (bodyStr) {
                     const body = JSON.parse(bodyStr);
                     const content = extractContent(body);
-                    const rawModel = match.modelExtractor(body);
+                    const rawModel = match.modelExtractor(body, url);
                     const originalModel = normalizeModel(rawModel);
                     const tokens = slash(content);
                     const originalInfo = getModel(originalModel);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "slash-tokens",
-  "version": "1.1.2",
+  "version": "1.2.0",
   "description": "Token Optimization for Context Engineers. 4.8 KB WASM. Sub-millisecond. Zero dependencies.",
   "main": "dist/index.js",
   "module": "dist/index.js",