NewApp Monitoring is liveSign up free
AgentbeamBeta
Open-source package vulnerability · PyPI

CVE-2026-29780: eml_parser: Path Traversal in Official Example Script Leads to Arbitrary File Write

mediumCVSS 3.1 5.5GHSA-389r-rccm-h3h5

Summary The official example script examples/recursivelyextractattachments.py contains a path traversal vulnerability that allows arbitrary file write outside the intended output directory. Attachment filenames extracted from parsed emails are directly used to construct output file paths without any sanitization, allowing an attacker-controlled filename to escape the target directory.

Published Mar 5, 2026 · Updated Mar 9, 2026

Affected packages

eml-parser

Ecosystem
PyPI
Affected
< 2.0.1
Fixed in
2.0.1
Check
pip show eml-parser

Advisory details

From the GitHub Advisory Database (CC BY 4.0).

Summary

The official example script examples/recursively_extract_attachments.py contains a path traversal vulnerability that allows arbitrary file write outside the intended output directory. Attachment filenames extracted from parsed emails are directly used to construct output file paths without any sanitization, allowing an attacker-controlled filename to escape the target directory.

Details

File: examples/recursively_extract_attachments.py Lines: 61–64

for a in m['attachment']:
    out_filepath = out_path / a['filename']  # No sanitization
    print(f'\tWriting attachment: {out_filepath}')
    with out_filepath.open('wb') as a_out:
        a_out.write(base64.b64decode(a['raw']))

The value a['filename'] is attacker-controlled via crafted email attachment headers:

Content-Disposition: attachment; filename="../outside/pwned.txt"

No path normalization or boundary validation is performed before writing.

PoC

  1. Create a malicious .eml file:
Content-Disposition: attachment; filename="../outside/pwned.txt"
  1. Run the example script:
python recursively_extract_attachments.py -p ./emails -o ./safe
  1. Expected: ./safe/pwned.txt
  2. Actual: ./outside/pwned.txt ← written outside the intended directory

Verified on Kali Linux with eml_parser installed via pip in a virtual environment.

Impact

This vulnerability is limited to the example script only and does not affect the core eml_parser library. However, as the script is part of the official repository and is likely to be adapted for production use, an attacker supplying a crafted email could achieve arbitrary file write within the execution context.

Potential attack scenarios include:

  • Cron job injection: filename="../../etc/cron.d/backdoor"
  • Web shell upload: filename="../../var/www/html/shell.php"
  • SSH key injection: filename="../../home/user/.ssh/authorized_keys"

Recommended Fix

import os.path

for a in m['attachment']:
    filename = os.path.basename(a['filename'])
    out_filepath = out_path / filename

    if not out_filepath.resolve().is_relative_to(out_path.resolve()):
        print(f'[!] Skipping suspicious filename: {a["filename"]}')
        continue

    print(f'\tWriting attachment: {out_filepath}')
    with out_filepath.open('wb') as a_out:
        a_out.write(base64.b64decode(a['raw']))

Credited to redyank.

References

How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology

More PyPI vulnerabilities

All PyPI entries