Read tool falsely reports non-encrypted PDF as password-protected

Bug Description

The Read tool falsely reports PDFs as "password-protected" when they are not encrypted. This has been observed repeatedly across multiple conversations.

Steps to Reproduce

1. Convert a PowerPoint (.pptx) file to PDF (standard export, no password set)
2. Use the Read tool to open the PDF:

``
Read(file_path="path/to/file.pdf", pages="1-5")
``

1. Error returned: PDF is password-protected. Please provide an unprotected version.

Verification

Using PyMuPDF to inspect the same file confirms it is not encrypted:

import fitz
doc = fitz.open("path/to/file.pdf")
print(f"Is encrypted: {doc.is_encrypted}")  # False
print(f"Needs pass: {doc.needs_pass}")       # 0

The file is fully readable via PyMuPDF — all 18 pages extract text successfully.

Workaround

Reading the PDF via PyMuPDF in the Bash tool works reliably as a fallback.

Environment

• Platform: Windows Server 2022 Datacenter (10.0.20348)
• Interface: VSCode extension
• PDF source: PowerPoint → PDF export (no encryption, no access restrictions)

Additional Context

This is a recurring issue — the user reports encountering it across multiple sessions. Sometimes retrying eventually works, but it is inconsistent.