Description

Budibase Server before 3.41.3 contains a server-side request forgery vulnerability in the query import endpoint that fails to validate user-supplied URLs before fetching content. Attackers can submit arbitrary URLs to retrieve responses from internal services including cloud metadata endpoints and other restricted network resources.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "7.1",
  "pubDate": "2026-08-28T12:16:36.360Z",
  "pubdate": "2026-08-28T12:16:36.360Z",
  "executiveSummary": "Budibase Server versions prior to 3.41.3 contain a critical server-side request forgery (SSRF) vulnerability within the query import functionality. This vulnerability stems from inadequate input validation of user-supplied URLs, allowing the application to fetch content from unauthorized locations.\nBy exploiting this flaw, a remote attacker can force the Budibase server to perform arbitrary HTTP requests to internal network resources. This capability facilitates reconnaissance of the internal infrastructure, potential exfiltration of sensitive configuration data, or interaction with internal services that are not exposed to the public internet.\nThe risk implication is significant, as the vulnerability can be leveraged to bypass network segmentation and access cloud metadata endpoints (e.g., AWS, GCP, or Azure IMDS), potentially leading to credential theft or broader compromise of the hosting environment.\nSuccessful exploitation requires the ability to interact with the query import endpoint. The vulnerability exposes the underlying host's internal network reachability, transforming the application into a proxy for malicious lateral movement.",
  "technicalDetails": "The root cause of this vulnerability lies in the insufficient sanitization and validation of destination parameters within the query import endpoint logic. The application fails to implement a blocklist or allowlist for requested URI schemes and domains, allowing the underlying request-handling library to process arbitrary URLs passed via user-supplied parameters.\nThe exploitation flow begins when an attacker submits a crafted payload containing a target URI to the query import function. Upon receiving this request, the Budibase server initiates an outbound network request on behalf of the attacker. Because the server does not enforce network-level restrictions or URL schema validation, it can be directed to interact with services running on localhost or within the internal VPC.\nA primary post-exploitation objective involves accessing cloud metadata service endpoints (e.g., http://169.254.169.254/). These endpoints often provide sensitive information, including IAM role credentials, security tokens, and host configuration details, without requiring additional authentication. By manipulating the query import feature, an attacker can retrieve these responses and exfiltrate them back to an external listener.\nFurthermore, the vulnerability allows for port scanning of the internal network infrastructure. By observing the time-based response differences or error messages returned by the query import feature, an attacker can identify active internal services, determine open ports, and interact with non-public APIs, databases, or management consoles that rely on network-based trust rather than authentication.\nThis vulnerability is present in all Budibase Server versions prior to 3.41.3. It does not require specific complex conditions to trigger, as the endpoint inherently performs a server-side fetch upon receiving the input. The impact is compounded in cloud-native environments where the server identity may be associated with highly privileged service accounts or roles, potentially leading to a total compromise of the associated cloud environment if metadata service responses are intercepted."
}