GHSA-CMCR-Q4JF-P6Q9

Vulnerability from github – Published: 2026-04-08 00:08 – Updated: 2026-04-08 00:08
VLAI?
Summary
WWBN AVideo has an Allowlisted downloadURL media extensions bypass SSRF protection and enable internal response exfiltration (Incomplete fix for CVE-2026-27732)
Details

Summary

The fix for CVE-2026-27732 is incomplete.

objects/aVideoEncoder.json.php still allows attacker-controlled downloadURL values with common media or archive extensions such as .mp4, .mp3, .zip, .jpg, .png, .gif, and .webm to bypass SSRF validation. The server then fetches the response and stores it as media content.

This allows an authenticated uploader to turn the upload-by-URL flow into a reliable SSRF response-exfiltration primitive.

Details

objects/aVideoEncoder.json.php accepts attacker-controlled downloadURL and passes it to downloadVideoFromDownloadURL().

Inside that function:

  1. the URL extension is extracted from the attacker-controlled path
  2. the extension is checked against an allowlist of normal encoder formats
  3. isSSRFSafeURL() is skipped for common media and archive extensions
  4. the URL is fetched via url_get_contents()
  5. the fetched body is written into video storage and exposed through normal media metadata

The current code still contains:

  • an extension-based bypass for SSRF validation
  • no mandatory initial-destination SSRF enforcement inside url_get_contents() itself

This means internal URLs such as:

http://127.0.0.1:9998/probe.mp4

remain reachable from the application host.

This issue is best described as an incomplete fix / patch bypass of CVE-2026-27732, not a separate unrelated SSRF class.

Proof of concept

  1. Log in as a low-privilege uploader.
  2. Start an HTTP service reachable only from inside the application environment, for example:
http://127.0.0.1:9998/probe.mp4
  1. Confirm that the service is not reachable externally.
  2. Send:
POST /objects/aVideoEncoder.json.php
downloadURL=http://127.0.0.1:9998/probe.mp4
format=mp4
  1. If needed, replay once against the returned videos_id with first_request=1 so the fetched bytes land in the normal media path.
  2. Query:
GET /objects/videos.json.php?showAll=1
  1. Recover videosURL.mp4.url.
  2. Download that media URL and observe that the body matches the internal-only response byte-for-byte.

Impact

An authenticated uploader can make the AVideo server fetch loopback or internal HTTP resources and persist the response as media content by supplying a downloadURL ending in an allowlisted extension such as .mp4, .jpg, .gif, or .zip. Because SSRF validation is skipped for those extensions, the fetched body is stored and later retrievable through the generated /videos/... media URL. Successful exploitation allows internal response exfiltration from private APIs, admin endpoints, or other internal services reachable from the application host.

Recommended fix

  • Apply isSSRFSafeURL() to all downloadURL inputs regardless of extension
  • Remove extension-based exceptions from SSRF enforcement
  • Move initial-destination SSRF validation into url_get_contents() so call sites cannot skip it
  • Avoid storing arbitrary fetched content directly as publicly retrievable media
  • Consider restricting upload-by-URL to an explicit allowlist of trusted fetch origins
Show details on source website

{
  "affected": [
    {
      "package": {
        "ecosystem": "Packagist",
        "name": "WWBN/AVideo"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "last_affected": "26.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-39370"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-04-08T00:08:47Z",
    "nvd_published_at": "2026-04-07T20:16:31Z",
    "severity": "HIGH"
  },
  "details": "## Summary\n\nThe fix for [CVE-2026-27732](https://github.com/WWBN/AVideo/security/advisories/GHSA-h39h-7cvg-q7j6) is incomplete.\n\n`objects/aVideoEncoder.json.php` still allows attacker-controlled `downloadURL` values with common media or archive extensions such as `.mp4`, `.mp3`, `.zip`, `.jpg`, `.png`, `.gif`, and `.webm` to bypass SSRF validation. The server then fetches the response and stores it as media content.\n\nThis allows an authenticated uploader to turn the upload-by-URL flow into a reliable SSRF response-exfiltration primitive.\n\n## Details\n\n`objects/aVideoEncoder.json.php` accepts attacker-controlled `downloadURL` and passes it to `downloadVideoFromDownloadURL()`.\n\nInside that function:\n\n1. the URL extension is extracted from the attacker-controlled path\n2. the extension is checked against an allowlist of normal encoder formats\n3. `isSSRFSafeURL()` is skipped for common media and archive extensions\n4. the URL is fetched via `url_get_contents()`\n5. the fetched body is written into video storage and exposed through normal media metadata\n\nThe current code still contains:\n\n- an extension-based bypass for SSRF validation\n- no mandatory initial-destination SSRF enforcement inside `url_get_contents()` itself\n\nThis means internal URLs such as:\n\n`http://127.0.0.1:9998/probe.mp4`\n\nremain reachable from the application host.\n\nThis issue is best described as an incomplete fix / patch bypass of `CVE-2026-27732`, not a separate unrelated SSRF class.\n\n## Proof of concept\n\n1. Log in as a low-privilege uploader.\n2. Start an HTTP service reachable only from inside the application environment, for example:\n\n```text\nhttp://127.0.0.1:9998/probe.mp4\n```\n\n3. Confirm that the service is not reachable externally.\n4. Send:\n\n```text\nPOST /objects/aVideoEncoder.json.php\ndownloadURL=http://127.0.0.1:9998/probe.mp4\nformat=mp4\n```\n\n5. If needed, replay once against the returned `videos_id` with `first_request=1` so the fetched bytes land in the normal media path.\n6. Query:\n\n```text\nGET /objects/videos.json.php?showAll=1\n```\n\n7. Recover `videosURL.mp4.url`.\n8. Download that media URL and observe that the body matches the internal-only response byte-for-byte.\n\n## Impact\n\nAn authenticated uploader can make the AVideo server fetch loopback or internal HTTP resources and persist the response as media content by supplying a `downloadURL` ending in an allowlisted extension such as `.mp4`, `.jpg`, `.gif`, or `.zip`. Because SSRF validation is skipped for those extensions, the fetched body is stored and later retrievable through the generated `/videos/...` media URL. Successful exploitation allows internal response exfiltration from private APIs, admin endpoints, or other internal services reachable from the application host.\n\n\n## Recommended fix\n\n- Apply `isSSRFSafeURL()` to all `downloadURL` inputs regardless of extension\n- Remove extension-based exceptions from SSRF enforcement\n- Move initial-destination SSRF validation into `url_get_contents()` so call sites cannot skip it\n- Avoid storing arbitrary fetched content directly as publicly retrievable media\n- Consider restricting upload-by-URL to an explicit allowlist of trusted fetch origins",
  "id": "GHSA-cmcr-q4jf-p6q9",
  "modified": "2026-04-08T00:08:47Z",
  "published": "2026-04-08T00:08:47Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/WWBN/AVideo/security/advisories/GHSA-cmcr-q4jf-p6q9"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-39370"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/WWBN/AVideo"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:L/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "WWBN AVideo has an Allowlisted downloadURL media extensions bypass SSRF protection and enable internal response exfiltration (Incomplete fix for CVE-2026-27732)"
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.


Loading…

Loading…

Loading…

Sightings

Author Source Type Date

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or observed by the user.
  • Confirmed: The vulnerability has been validated from an analyst's perspective.
  • Published Proof of Concept: A public proof of concept is available for this vulnerability.
  • Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
  • Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
  • Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
  • Not confirmed: The user expressed doubt about the validity of the vulnerability.
  • Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.


Loading…

Detection rules are retrieved from Rulezet.

Loading…

Loading…