Hash Generator

Hash text with SHA-256, SHA-384, SHA-512, or the legacy SHA-1 and MD5 algorithms.

Use this tool

MD5 is cryptographically broken and must not be used for new password storage or security-sensitive authentication. It is available here for legacy compatibility, checksums, interoperability, and migration/testing of existing systems. SHA-1 is collision-broken and is kept for the same kind of legacy use, not for new signatures or authentication.

Result

The result appears in this panel after you run the tool.

About this tool

SHA-256, SHA-384, and SHA-512 are computed with Web Crypto. SHA-1 is also Web Crypto, and it is grouped with MD5 because both are legacy. MD5 is not part of Web Crypto, so this page loads a small MD5 implementation only when you choose MD5.

MD5 is cryptographically broken and must not be used for new password storage or security-sensitive authentication. It is available here for legacy compatibility, checksums, interoperability, and migration/testing of existing systems. SHA-1 is collision-broken. Do not use it for new signatures or authentication. Neither algorithm is described as secure on this page.

How to use it

  1. Paste the text.
  2. Choose the algorithm. SHA-256 is the default for a new hash.
  3. Hash the text. MD5 is loaded only if you choose MD5.

Limitations

A digest of UTF-8, with legacy algorithms labeled

The three SHA-2 choices

The default algorithm is SHA-256, with SHA-384 and SHA-512 beside it. Those three call crypto.subtle.digest on the UTF-8 bytes of the textarea. The panel prints the algorithm name and the lowercase hex digest. There is no key. The same text always produces the same digest, which is what you want when you are comparing a paste to a published fingerprint, and it is what you do not want from a password store. A digest is not a reversible encoding. You cannot turn the hex back into the paragraph. It is not a ciphertext either: nothing is hidden from someone who already has the paragraph, because they can recompute the digest.

Use these three when you are creating a new fingerprint of text. Changing one character changes the digest. A different Unicode normalization, or a different newline, is a different byte string and will not match. The page does not normalize.

SHA-1 and MD5 stay available for old checksums

SHA-1 is still computed with Web Crypto, and the result appends a notice that collisions are practical and that the algorithm should not back new signatures or authentication. MD5 is not in Web Crypto. Choosing it loads a small legacy implementation and hashes the same UTF-8 bytes. The result line carries a notice that MD5 is broken for new password storage and for authentication. Both algorithms remain on the menu so you can match a checksum an older system already published, or compare a migration, or test an interoperability case. Matching a legacy digest is the job. Calling either one secure is not. Do not put new passwords through them. Do not accept them as proof against an attacker who can craft collisions. SHA-1’s notice is about collisions and signatures. MD5’s notice is about password storage and authentication. Read the line the page actually appends for the algorithm you picked.

What this page is not

It is not HMAC. Without a key, anyone can recompute the digest, so it does not authenticate a sender. It is not a file hash: what gets hashed is the paste’s UTF-8, not a raw file. A file that contains a BOM or CRLF will not match a paste of the visible text. The file checksum page reads file bytes and does not offer MD5 or SHA-1. It is not CRC-32 either. CRC is for accidental change on a different page. Encoding the text as base64 or percent escapes would still be readable. Hashing it is a one-way fingerprint, and the legacy choices are fingerprints you should only be comparing, not deploying.

Privacy

Your input is processed in your browser and is not uploaded to UtilChain. File names, text, and results are not sent to analytics.