TXT · Text & Data tools

Hash Extractor (MD5, SHA-1, SHA-256 from Text)

Input
Hashes
0hashes0lines scanned
Go to full Text Pattern Extractor

Shareable link

Settings are written to the URL as you change them. Nothing differs from the defaults yet.

Extract hashes from a report

Paste a threat report, a SHA256SUMS file, a package lock or a malware sandbox export. Hashes are grouped under MD5, SHA-1, SHA-256 and SHA-512 headings, each one listed once in lowercase, in the order it first appears. Upper- and lowercase copies of the same hash count as one.

Grouped by length, not identified

A hash carries no label. The grouping comes from length alone: 32 hex characters is the size of an MD5 digest, 40 of SHA-1, 64 of SHA-256, 128 of SHA-512. The same 32 characters could be an NTLM hash or a UUID written without dashes, and 64 could be SHA3-256 or an Ethereum transaction hash. Treat the headings as the likely algorithm and confirm against the source.

What is skipped

A hex run must stand alone. It cannot be one character too long or too short, and it cannot be glued to other letters, which rules out the hex-looking stretches inside Base64 data. Dashed UUIDs, all-digit strings and bcrypt or Argon2 strings (which carry their own $2b$ or $argon2id$ prefix and are not hex) are not listed.

Frequently Asked Questions

Only by length. Exactly 32 hex characters is grouped as MD5, 40 as SHA-1, 64 as SHA-256 and 128 as SHA-512. Other algorithms share those lengths (NTLM is 32, SHA3-256 and Keccak-256 are 64), so each heading says by length.

It must be exactly 32, 40, 64 or 128 hex characters and must not touch other letters or digits, so a 33-character run, a slice of a Base64 blob or a dashed UUID is skipped. A run of digits only is skipped too, because a 32-digit number is far more often an ID.

Yes. A hash in a path such as /file/<sha256>/detection or after a label such as sha256= is found, and a 0x prefix is dropped.

Explore Our Tools

Browse all tools