Description

A vulnerability has been found in Weaviate Verba up to 2.1.3. Affected by this vulnerability is the function get_environment of the file goldenverba/components/util.py of the component generate_stream Endpoint. The manipulation leads to information disclosure. It is possible to initiate the attack remotely. The exploit has been disclosed to the public and may be used. The vendor was contacted early about this disclosure but did not respond in any way.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "5.3",
  "pubDate": "2026-10-04T11:16:31.197Z",
  "pubdate": "2026-10-04T11:16:31.197Z",
  "executiveSummary": "A critical information disclosure vulnerability exists in Weaviate Verba up to version 2.1.3 within the 'generate_stream' endpoint.\nThe vulnerability originates from the 'get_environment' function located in 'goldenverba/components/util.py', which improperly exposes sensitive environment-level data.\nSuccessful exploitation allows remote, unauthenticated attackers to retrieve system configuration details and environment variables.\nThe risk is significant as it may leak API keys, database credentials, or internal configuration paths, providing an attacker with a foothold for further exploitation or lateral movement within the infrastructure.\nThe vulnerability is remotely exploitable, requiring no prior authentication, and the exploit is publicly available, increasing the threat level for deployments that remain unpatched or unhardened.\nVendor non-responsiveness regarding this disclosure necessitates immediate independent mitigation by administrators.",
  "technicalDetails": "The vulnerability resides in the 'get_environment' function within the 'goldenverba/components/util.py' module of the Weaviate Verba application.\nThis function is invoked by the 'generate_stream' endpoint to retrieve and return environmental configuration data to the requester.\nThe root cause is an insecure implementation that aggregates environment variables and potentially sensitive system-wide configuration metadata, returning this data directly in the HTTP response body without adequate sanitization, authorization, or filtering.\nThe attack flow involves a remote attacker sending a crafted request to the 'generate_stream' endpoint. Because the application endpoint fails to validate the request origin or user privileges, it executes the 'get_environment' function and serializes the resulting object into the response stream.\nThis output typically includes sensitive environment variables (such as 'OPENAI_API_KEY', 'WEAVIATE_URL', or 'GOLDEN_VERBA_API_KEY') and other runtime configuration settings defined in the application's process environment.\nSince the vulnerability is exposed via an API endpoint, it is accessible via the network without requiring any authentication tokens or session identifiers. An attacker can use common tools like 'curl' or 'HTTPie' to interact with the 'generate_stream' endpoint, successfully triggering the disclosure of sensitive information.\nThe impact of this disclosure is high, as the leaked variables often contain credentials necessary to access secondary services (e.g., LLM providers, Weaviate vector databases, or cloud infrastructure). This constitutes a classic information leakage pattern where the application architecture inadvertently serves as a source for reconnaissance, enabling attackers to escalate their access or facilitate data exfiltration.\nGiven that the exploit is public, the attack surface is effectively open, and automated scanning by adversaries is a likely occurrence. The lack of patching means that the endpoint continues to act as a vector for credential theft as long as the affected function is reachable via the public-facing 'generate_stream' endpoint."
}