Unlock PDF

Make an unlocked copy in three steps

1. Choose one local PDF up to 20 MiB and 500 pages. Use only documents you are authorized to unlock; possession of a password does not establish ownership.
2. Enter the existing password exactly, including case and ordinary spaces, or leave it empty for a supported permissions-only document. Select Unlock PDF and wait for authentication, preservation and output checks.
3. Download unlocked.pdf only after the verified result appears. Anyone with that copy can open it. Keep the protected original if you still need its protection or history.

What unlocking changes

A supported encrypted document becomes a separate unencrypted copy. Existing user or owner authentication is accepted; an empty credential can be valid for permissions-only encryption. This is not password guessing or recovery.

The tool preserves admitted reachable pages, resources, ordinary forms and annotations, navigation, tags, layers, metadata and attachments. It does not flatten pages, substitute fonts or silently strip a feature to make the file pass.

Existing password handling differs from creating a password. The raw draft stays unchanged; bounded legacy encoding and R6 reader-compatibility candidates may be tried. This does not guarantee every producer or Unicode password combination will open.

The complete candidate must pass structural, payload and independent-reader checks without a password before Download is enabled. A successful first page alone does not establish that an attachment or the rest of the file is unencrypted.

Common outcomes

An admitted eight-page protected report becomes a verified eight-page unencrypted copy. Its result shows page count and actual output size, never the password or an invented amount of space saved.

A supported permissions-only PDF may accept an empty credential and produce an unlocked copy. A valid PDF that is already unencrypted instead reports that no new copy is needed; it is not rewritten just to offer a download.

An actual signature, XFA content, unsupported encryption, incompatible credential/reader combination or failed preservation check stops the operation. The tool does not silently flatten or strip the document and call that success.

Important document and credential limits

Only the documented canonical Standard encryption families are admitted. Custom handlers, special stream or attachment crypt filters and malformed or recovered documents are rejected. Reading old RC4 protection does not create new weak encryption.

Actual signatures, certification and usage-rights structures and XFA are not supported. Ordinary unsigned signature fields can remain. Conformance metadata is retained, but no PDF/A, PDF/X or cryptographic signature validation is certified.

Passwords are not trimmed, and new-password strength or a 127-byte input rule is not imposed. The original encryption algorithm still has its own byte behavior. Unicode compatibility attempts are reader handling, not universal or strict Unicode-3.2 SASLprep support.

Input is capped at 20 MiB and 500 pages, output at 40 MiB, inspection at 45 seconds and the whole operation at 120 seconds. Object, graph, scalar, stream and attachment budgets also apply. These are rejection limits, not a guarantee that every file below them will complete or fit memory.

Unlocking is not redaction or secure erasure. Attachments and embedded encrypted PDFs remain content, not recursively unlocked files. File bytes, object numbers, orphan objects and old incremental history are not preserved as an archival or forensic copy.

Local processing without saved passwords

The tool adds no document upload, saved password, history or input-bearing share link. Download happens only when requested. Workers and transient data are released best-effort, but browser memory, extensions, existing site services and files you keep are not covered by an erasure or universal security guarantee.