diff --git a/README.md b/README.md
index 13fa1b6..1878dd8 100644
--- a/README.md
+++ b/README.md
@@ -92,7 +92,7 @@ const data = createTar(
 
 ### Compression
 
-You can optionaly use `createTarGzip` or `createTarGzipStream` to create a compressed tar data stream (returned value is a [`Promise<Uint8Array>`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Uint8Array) or [`RedableStream`](https://developer.mozilla.org/en-US/docs/Web/API/ReadableStream) piped to [`CompressionStream`](https://developer.mozilla.org/en-US/docs/Web/API/CompressionStream))
+You can optionally use `createTarGzip` or `createTarGzipStream` to create a compressed tar data stream (returned value is a [`Promise<Uint8Array>`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Uint8Array) or [`ReadableStream`](https://developer.mozilla.org/en-US/docs/Web/API/ReadableStream) piped to [`CompressionStream`](https://developer.mozilla.org/en-US/docs/Web/API/CompressionStream))
 
 ```js
 import { createTarGzip, createTarGzipStream } from "nanotar";
@@ -139,6 +139,14 @@ const files = parseTar(data);
 
 Parsed files array has two additional properties: `size` file size and `text`, a lazy getter that decodes `data` view as a string.
 
+You can filter iterms to read using `filter` option:
+
+```ts
+const files = parseTar(data, {
+  filter: (file) => file.name.starsWith("dir/"),
+});
+```
+
 ### Decompression
 
 If input is compressed, you can use `parseTarGzip` utility instead to parse it (it used [`DecompressionStream`](https://developer.mozilla.org/en-US/docs/Web/API/DecompressionStream) internally and return a `Promise<Uint8Array>` value)
diff --git a/src/parse.ts b/src/parse.ts
index ab34e9a..5dc4b00 100644
--- a/src/parse.ts
+++ b/src/parse.ts
@@ -1,19 +1,31 @@
-import type { ParsedTarFileItem } from "./types";
+import type { ParsedTarFileItem, ParsedTarFileItemMeta } from "./types";
 
 const TAR_TYPE_FILE = 0;
 const TAR_TYPE_DIR = 5;
 
+export interface ParseTarOptions {
+  /**
+   * A filter function that determines whether a file entry should be skipped or not.
+   */
+  filter?: (file: ParsedTarFileItemMeta) => boolean;
+}
+
 /**
  * Parses a TAR file from a binary buffer and returns an array of {@link TarFileItem} objects.
  *
  * @param {ArrayBuffer | Uint8Array} data - The binary data of the TAR file.
  * @returns {ParsedTarFileItem[]} An array of file items contained in the TAR file.
  */
-export function parseTar(data: ArrayBuffer | Uint8Array): ParsedTarFileItem[] {
+export function parseTar(
+  data: ArrayBuffer | Uint8Array,
+  opts?: ParseTarOptions,
+): ParsedTarFileItem[] {
   const buffer = (data as Uint8Array).buffer || data;
 
   const files: ParsedTarFileItem[] = [];
 
+  const filter = opts?.filter;
+
   let offset = 0;
 
   while (offset < buffer.byteLength - 512) {
@@ -35,6 +47,9 @@ export function parseTar(data: ArrayBuffer | Uint8Array): ParsedTarFileItem[] {
     // File size (offset: 124 - length: 12)
     const size = _readNumber(buffer, offset + 124, 12);
 
+    // Calculate next seek offset based on size
+    const seek = 512 + 512 * Math.trunc(size / 512) + (size % 512 ? 512 : 0);
+
     // File mtime (offset: 136 - length: 12)
     const mtime = _readNumber(buffer, offset + 136, 12);
 
@@ -54,20 +69,11 @@ export function parseTar(data: ArrayBuffer | Uint8Array): ParsedTarFileItem[] {
     // File owner group (offset: 297 - length: 32)
     const group = _readString(buffer, offset + 297, 32);
 
-    // File data (offset: 512 - length: size)
-    const data =
-      _type === TAR_TYPE_DIR
-        ? undefined
-        : new Uint8Array(buffer, offset + 512, size);
-
-    files.push({
+    // Group all file metadata
+    const meta: ParsedTarFileItemMeta = {
       name,
       type,
       size,
-      data,
-      get text() {
-        return new TextDecoder().decode(this.data);
-      },
       attrs: {
         mode,
         uid,
@@ -76,12 +82,29 @@ export function parseTar(data: ArrayBuffer | Uint8Array): ParsedTarFileItem[] {
         user,
         group,
       },
-    });
+    };
 
-    offset += 512 + 512 * Math.trunc(size / 512);
-    if (size % 512) {
-      offset += 512;
+    // Filter
+    if (filter && !filter(meta)) {
+      offset += seek;
+      continue;
     }
+
+    // File data (offset: 512 - length: size)
+    const data =
+      _type === TAR_TYPE_DIR
+        ? undefined
+        : new Uint8Array(buffer, offset + 512, size);
+
+    files.push({
+      ...meta,
+      data,
+      get text() {
+        return new TextDecoder().decode(this.data);
+      },
+    });
+
+    offset += seek;
   }
 
   return files;
@@ -98,7 +121,7 @@ export function parseTar(data: ArrayBuffer | Uint8Array): ParsedTarFileItem[] {
  */
 export async function parseTarGzip(
   data: ArrayBuffer | Uint8Array,
-  opts: { compression?: CompressionFormat } = {},
+  opts: ParseTarOptions & { compression?: CompressionFormat } = {},
 ): Promise<ParsedTarFileItem[]> {
   const stream = new ReadableStream({
     start(controller) {
@@ -109,7 +132,7 @@ export async function parseTarGzip(
 
   const decompressedData = await new Response(stream).arrayBuffer();
 
-  return parseTar(decompressedData);
+  return parseTar(decompressedData, opts);
 }
 
 function _readString(buffer: ArrayBufferLike, offset: number, size: number) {
diff --git a/src/types.ts b/src/types.ts
index 4123240..8060be0 100644
--- a/src/types.ts
+++ b/src/types.ts
@@ -34,6 +34,8 @@ export interface ParsedTarFileItem extends TarFileItem {
   readonly text: string;
 }
 
+export type ParsedTarFileItemMeta = Omit<ParsedTarFileItem, "data" | "text">;
+
 export interface TarFileAttrs {
   /**
    * File mode in octal (e.g., `664`) represents read, write, and execute permissions for the owner, group, and others.
diff --git a/test/index.test.ts b/test/index.test.ts
index 676c23d..387356e 100644
--- a/test/index.test.ts
+++ b/test/index.test.ts
@@ -92,4 +92,17 @@ describe("nanotar", () => {
       ]
     `);
   });
+
+  it("parseTar (with filter)", async () => {
+    const data = await createTarGzip(fixture);
+    const files = (
+      await parseTarGzip(data, {
+        filter: (file) => file.name.startsWith("foo/"),
+      })
+    ).map((f) => ({
+      ...f,
+      data: f.data ? inspect(f.data).replace(/\s+/g, " ") : undefined,
+    }));
+    expect(files.map((f) => f.name)).toMatchObject(["foo/bar.txt"]);
+  });
 });