UTF-8 Base64 & Offline JWT Decoder

JWT Decode

Encode or decode Base64 and Base64URL with full Unicode/CJK safety—and automatically decode JWT headers and claims with 0 bytes uploaded.

100% Client-Side — Verify 0 bytes uploaded in Network tabVerify 0 bytes sent in Network tab

or drop a text or data file here · Ctrl V to paste · or drop .txt, .json, .jwt · Ctrl V to paste Base64 or JWT
  • 0 bytes uploaded — verify offline
  • Unicode & CJK UTF-8 safe
  • Auto-decodes JWT claims

Instant Before & After Preview

Auto-decodes JWT claims
Before (Input)sample.jwt
eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiJ1c2VyXzEyMyIsIm5hbWUiOiJNYXlhIExpbiIsImxvY2FsIjoidHJ1ZSJ9.dGVzdF9zaWduYXR1cmU
After (Output).txt
{
  "header": {
    "alg": "HS256",
    "typ": "JWT"
  },
  "payload": {
    "sub": "user_123",
    "name": "Maya Lin",
    "local": "true"
  },
  "signature": "dGVzdF9zaWduYXR1cmU"
}

Source vs. target format specification comparison: Plain Text (TXT) → Base64

PropertySource: Plain Text (TXT)Target: Base64
Structure typePlain TextPlain Text
Human readabilityHighLow
Nesting supportNoNo
LLM token efficiencyHighLow
File extension.txt.txt
MIME typetext/plaintext/plain
Best forUniversal logs, raw transcripts, cleaned LLM prompts, scripts, and zero-markup reading.JWT token inspection, HTTP Basic Auth headers, Kubernetes Secret manifests, and URL-safe binary tokens.

How it works

How to encode Base64 and inspect JWTs safely

Decode production auth tokens and encode UTF-8 strings directly in browser RAM without touching external token debuggers.

  1. Paste JWT or Base64 textPaste a 3-part JWT (eyJ...), Base64URL string, or plain UTF-8 text.
  2. Select Encode or DecodeKeep Auto-decode JWT and URL-safe Base64 enabled to decode headers and JSON claims automatically.
  3. Inspect decoded JSONReview formatted JWT header/payload claims or copy clean Base64 output.

FAQ

Frequently asked questions

Is it safe to paste real production JWTs into TxtTo?

Yes. Everything runs 100% locally in your browser's JavaScript memory with zero network uploads or token logging.

Why doesn't TxtTo crash on Chinese, Japanese, or emoji characters like browser btoa()?

Native window.btoa() only handles Latin-1 characters. TxtTo converts through UTF-8 TextEncoder/TextDecoder byte arrays first so all Unicode text encodes cleanly.

Does URL-safe Base64 omit trailing = padding and use - / _?

Yes. Toggling URL-safe Base64 swaps + and / for - and _ and handles missing Base64URL padding gracefully during decoding.