QQuickToolz
All tools

JWT Decoder

Paste a JSON Web Token to inspect its header, payload and signature. Decode only — the signature is not verified and nothing leaves your browser.

Paste a JWT above — decoding happens locally, nothing is sent anywhere.

How it works

JWT Decoder unpacks a JSON Web Token into its three parts — header, payload and signature — and shows them as readable JSON. Time claims (iat, nbf, exp) are turned into human dates with an expired/valid verdict. Everything is decoded locally; the token is never sent anywhere and the signature is not verified.

Quick start

  1. Paste the token (a leading 'Bearer ' is tolerated).
  2. Read the header, payload and claim timeline.
  3. Copy the payload JSON if you need it elsewhere.

Frequently asked questions

Does the tool verify the signature?

No — this is a decoder only. Signature verification requires the secret or public key, which we don't have and don't ask for.

Is my token uploaded?

No. Decoding happens entirely in your browser.

What are exp, iat and nbf?

Standard claims: exp says when the token expires, iat when it was issued, nbf when it becomes valid.

Why can I see the payload without the signature key?

The payload is only Base64URL-encoded, not encrypted. Anyone holding a token can read it — don't put secrets inside a JWT.

Anatomy of a JWT, in five minutes

Three Base64URL segments, zero encryption — what a token actually says about you. · 5 min read

Three segments, each a different job

A JSON Web Token is three pieces of text joined by dots: header.payload.signature. The header names the signing algorithm and token type. The payload carries the claims — the statements being made. The signature is the cryptographic glue that lets a server prove those claims came from the issuer and were not modified in transit.

Each segment is Base64URL-encoded JSON: standard Base64 with + and / swapped for - and _ and padding trimmed, so the token survives inside URLs, headers and cookies without escaping. Decoding is therefore trivial — which is exactly why you can read a token's contents without any key.

Signed, not encrypted — the distinction that matters

The signature gives integrity and authenticity: nobody can change 'role: user' to 'role: admin' without invalidating it, assuming the key stayed secret. But nothing is hidden. Anyone holding the token — a teammate, a proxy log, a browser extension — can read every claim inside it. Putting a password, a private key or a national ID number into a JWT payload is a real, recurring security incident pattern.

The signature also does not say anything about whether the claims are true. It proves who issued the token, not that the issuer was right to do so. Revocation is the other classic gap: a stateless JWT remains valid until exp even if the user was just banned, unless the server adds its own revocation list or keeps sessions short.

The registered claims worth knowing

A handful of claim names have standard meanings: iss (issuer), sub (subject — usually a user id), aud (audience), iat (issued at), nbf (not before) and exp (expiry). Servers are expected to check exp and aud; clocks skew, so real systems allow a small leeway of a minute or two.

exp is a Unix timestamp in seconds — not milliseconds. One of the most common integration bugs is a client computing expiry in milliseconds and getting a token that 'never' expires. Reading the exp value with a decoder like this one — and checking it against a real clock — takes ten seconds and settles the argument.

How to inspect a token responsibly

Decoding tools (this one included) parse the segments locally, so the token is not transmitted anywhere. Still, treat tokens like credentials: prefer pasting a test or expired token, or redact sensitive claims first. What you usually need from an inspection is right at the top anyway — the algorithm, the expiry, and the audience the token claims to be for.