csvql-query 2.3.0 → 2.4.1

package/README.md

@@ -2,7 +2,7 @@
 
 Fast SQL queries on CSV files from Node.js — powered by a Zig/SIMD engine via N-API.
 
-The engine streams through the file instead of loading it into memory, so RAM stays ~5 MB regardless of file size. Prebuilt native binaries ship for macOS (arm64/x64), Linux (x64/arm64), and Windows (x64) — no compiler needed.
+The engine streams through the file instead of loading it into memory, so RAM stays flat — tens of MB — regardless of file size. Prebuilt native binaries ship for macOS (arm64/x64), Linux (x64/arm64), and Windows (x64) — no compiler needed.
 
 ## Install
 

package/index.js

@@ -4,8 +4,8 @@
  * csvql — fast SQL queries on CSV files from Node.js
  *
  * Uses the csvql Zig/SIMD engine via N-API. The engine streams through the
- * file without loading it into memory, so RAM usage stays ~5 MB regardless
- * of file size.
+ * file instead of loading it into memory, so RAM usage stays flat (tens of
+ * MB) regardless of file size.
  *
  * @example
  * const csvql = require('csvql-query');
@@ -168,7 +168,7 @@ function cleanUp(temps) {
  * as an array of objects.
  *
  * The engine streams through the file without loading all rows into memory,
- * so RAM stays ~5 MB for any file size. Values are typed where possible
+ * so RAM stays flat (tens of MB) for any file size. Values are typed where possible
  * (numbers come back as JS numbers via JSON.parse).
  *
  * @param {string}       sql   SQL query. File paths go in FROM/JOIN clauses.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "csvql-query",
-  "version": "2.3.0",
+  "version": "2.4.1",
   "description": "Fast SQL queries on CSV files — Zig/SIMD engine via N-API",
   "main": "index.js",
   "types": "index.d.ts",
@@ -21,7 +21,10 @@
     "simd",
     "fast",
     "etl",
-    "parser"
+    "parser",
+    "agent",
+    "ai",
+    "mcp"
   ],
   "license": "MIT",
   "repository": {