squeezr-ai 1.10.5 → 1.10.6

package/README.md

@@ -103,6 +103,8 @@ localhost:8080  (Squeezr proxy)
 Your provider's API (Anthropic / OpenAI / Google / Ollama)
 ```
 
+**MCP tool results are compressed automatically.** Any tool result that passes through the proxy — including results from MCP servers (Linear, GitHub, Slack, planning tools, custom MCPs) — goes through the same compression pipeline. No configuration needed; Squeezr treats MCP tool results identically to built-in tools. In practice MCP responses are often large JSON payloads that compress 70-94%.
+
 Recent content is always preserved untouched — by default the last 3 tool results are never compressed. Your CLI always has full context for what it's currently working on.
 
 ---

package/dist/gain.js

@@ -27,21 +27,30 @@ const savedChars = data.total_saved_chars;
 const originalChars = data.total_original_chars;
 const CHARS_PER_TOKEN = 3.5;
 const savedTokens = Math.round(savedChars / CHARS_PER_TOKEN);
-const pct = originalChars > 0 ? Math.round((savedChars / originalChars) * 1000) / 10 : 0;
 const byTool = (data.by_tool ?? {});
+// Savings % on tool results only (what Squeezr actually compresses).
+// Using total context chars gives a misleadingly low number because the
+// denominator includes user messages, Claude responses, and history
+// re-sent on every request — none of which Squeezr touches.
+const toolOriginal = Object.values(byTool).reduce((s, d) => s + d.originalChars, 0);
+const toolSaved = Object.values(byTool).reduce((s, d) => s + d.savedChars, 0);
+const toolPct = toolOriginal > 0 ? Math.round((toolSaved / toolOriginal) * 1000) / 10 : 0;
+// Context reduction: how much smaller the overall payload became
+const ctxPct = originalChars > 0 ? Math.round((savedChars / originalChars) * 1000) / 10 : 0;
 console.log('┌─────────────────────────────────────────┐');
 console.log('│          Squeezr — Token Savings         │');
 console.log('├─────────────────────────────────────────┤');
-console.log(`│  Requests      ${String(requests).padEnd(26)}│`);
-console.log(`│  Saved chars   ${String(savedChars.toLocaleString()).padEnd(26)}│`);
-console.log(`│  Saved tokens  ${String(savedTokens.toLocaleString()).padEnd(26)}│`);
-console.log(`│  Savings       ${String(`${pct}%`).padEnd(26)}│`);
+console.log(`│  Requests         ${String(requests).padEnd(23)}│`);
+console.log(`│  Saved chars      ${String(savedChars.toLocaleString()).padEnd(23)}│`);
+console.log(`│  Saved tokens     ${String(savedTokens.toLocaleString()).padEnd(23)}│`);
+console.log(`│  Tool savings     ${String(`${toolPct}%`).padEnd(23)}│`);
+console.log(`│  Context reduction ${String(`${ctxPct}%`).padEnd(22)}│`);
 if (Object.keys(byTool).length > 0) {
     console.log('├─────────────────────────────────────────┤');
     console.log('│  By Tool                                 │');
-    for (const [tool, d] of Object.entries(byTool)) {
-        const toolPct = d.originalChars > 0 ? Math.round((d.savedChars / d.originalChars) * 1000) / 10 : 0;
-        const line = `  ${tool} (${d.count}x): -${toolPct}%`;
+    for (const [tool, d] of Object.entries(byTool).sort((a, b) => b[1].savedChars - a[1].savedChars)) {
+        const pct = d.originalChars > 0 ? Math.round((d.savedChars / d.originalChars) * 1000) / 10 : 0;
+        const line = `  ${tool} (${d.count}x): -${pct}%`;
         console.log(`│${line.padEnd(41)}│`);
     }
 }

package/dist/version.d.ts

@@ -1 +1 @@
-export declare const VERSION = "1.10.5";
+export declare const VERSION = "1.10.6";

package/dist/version.js

@@ -1 +1 @@
-export const VERSION = '1.10.5';
+export const VERSION = '1.10.6';

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "squeezr-ai",
-  "version": "1.10.5",
+  "version": "1.10.6",
   "description": "AI proxy that compresses Claude Code, Codex, Aider, Gemini CLI and Ollama context windows to save thousands of tokens per session",
   "keywords": [
     "claude",