Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessLanguage server initiates outbound HTTP connection
-
DeliveryNo workspace http.proxy setting found
-
ExploitRaw HTTP_PROXY env var with embedded credentials read
-
ExecutionFull URL logged at INFO level without redaction
-
PersistLog file read by local user or shared externally
-
ImpactProxy credentials extracted and reused
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A09:2025 - Security Logging & Alerting Failures
ASVS 5.0
V16 - Security Logging and Error Handling Log data sanitization
|
| Exploitation Conditions | Two specific conditions must both be true for credential exposure to occur: (1) the affected user must have a proxy URL with embedded Basic-auth credentials set in one of the HTTP_PROXY, HTTPS_PROXY, http_proxy, or https_proxy environment variables, and (2) no explicit http.proxy setting must be configured at the IDE workspace level-because a workspace setting causes the language server to bypass the environment variable lookup entirely. Exploitation of the resulting log content additionally requires that a second party gain read access to the log file, either through shared local file system access (PR:L per CVSS) or through the common practice of attaching logs to bug reports. The vulnerability does NOT trigger in proxy environments using NTLM, Kerberos, or other auth schemes where credentials are not embedded in the URL string. It is also inert for users whose proxies require no authentication, or who explicitly configure the workspace proxy setting. |
| Impact | The vendor-assigned CVSS 3.1 base score of 3.3 (AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:N/A:N) is consistent with the threat model: exploitation requires local access, low-privilege access to the log file, and the specific configuration condition where no workspace proxy setting overrides the environment variable fallback. There is no EPSS score provided, no CISA KEV listing, and no public exploit code identified, making this a low-urgency finding in absolute terms. However, the real-world risk is not trivial: in enterprise environments, proxy credentials may grant network-level access to internal resources, and log files are routinely shared in support tickets, GitHub issues, or bug reports-widening the exposure beyond a strictly local threat. The CVSS local attack vector accurately reflects that an attacker must have existing system access, but it does not capture the secondary exposure vector of log-sharing. Organizations with strict secrets management or compliance requirements (PCI-DSS, SOC 2) may assess this higher than the raw score suggests. No CVSS signal conflicts are present; the vector is internally consistent with the description. |
| Exploit ScenarioDerived | A developer working on a corporate network has HTTP_PROXY set to http://jdoe:s3cr3t@proxy.corp.example.com:8080 in their environment. They open a Spring project in VS Code with no workspace proxy setting configured. The Spring Tools language server starts, makes an outbound HTTP request, and logs the full proxy URL at INFO level to the language server log. The developer later files a GitHub issue reproducing a build problem and attaches the log-publicly exposing jdoe's proxy credentials to anyone who views the issue. No exploit code is needed; the log contents are the artifact of exploitation. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Jul 30, 2026
Spring Tools language server exposes proxy Basic-auth credentials in cleartext log files, affecting Spring Tools for Eclipse 5.2.0 and earlier and Spring Tools for VSCode/Cursor/Theia 2.2.0 and earlier. When no explicit http.proxy workspace setting is configured, the language server reads the HTTP_PROXY/HTTPS_PROXY (and lowercase variants) environment variables and writes their raw values-including embedded credentials in the form http://user:pass@proxy:8080-to its log at INFO level without redaction. Because language server logs are often shared in bug reports or are readable by other local users and processes, this results in proxy credential disclosure. No public exploit has been identified at time of analysis, and CISA KEV listing is absent.
RemediationAI
Consult the official Spring security advisory at https://spring.io/security/cve-2026-59326 for confirmed fixed versions and upgrade Spring Tools for Eclipse beyond 5.2.0 and Spring Tools for VSCode/Cursor/Theia beyond 2.2.0 as soon as vendor-released patched builds are available; exact fix versions are not independently confirmed in the data available at time of this analysis. As an immediate workaround, configure an explicit http.proxy workspace setting within the IDE rather than relying on environment variable fallback-this prevents the language server from reading and logging the raw environment variable value. Alternatively, avoid embedding credentials directly in proxy environment variables by using proxy configurations that rely on OS-level credential stores or NTLM/Kerberos authentication where the URL itself contains no embedded secret. If neither workaround is feasible, restrict read permissions on language server log files to the owning user only (chmod 600 on Linux/macOS) and establish a policy against attaching raw log files to bug reports without credential scrubbing.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 2 packages in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Spring Tools (formerly Spring Tool Suite) is a set of language server extensions and plugins that provide Spring Framework development support inside Eclipse, VS Code, Cursor, and Theia IDEs. Language servers run as background processes and communicate with IDEs via the Language Server Protocol (LSP). When the language server needs to make outbound HTTP calls (e.g., to fetch metadata or resolve dependencies), it performs a runtime check: if no workspace-level http.proxy setting is present, it falls back to reading OS-level proxy environment variables (HTTP_PROXY, HTTPS_PROXY, http_proxy, https_proxy). The raw string value-which corporate environments routinely encode as http://username:password@proxyhost:port-is then emitted at INFO log verbosity. INFO is a default-enabled log level, meaning this behavior is active in standard deployments with no special debug configuration required. Although no CWE is formally assigned in the NVD entry, this is substantively CWE-532 (Insertion of Sensitive Information into Log File). Affected CPEs confirmed by vendor: cpe:2.3:a:spring:spring_tools_for_eclipse:*:*:*:*:*:*:*:* (≤5.2.0) and cpe:2.3:a:spring:spring_tools_for_vscode_/_cursor_/_theia:*:*:*:*:*:*:*:* (≤2.2.0).
Know when Spring gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:N/A:N
Local vector confirmed by log-file access model; PR:L because a standard local user account is sufficient; C:L because only proxy credentials are exposed, not broader system confidentiality.
Primary rating from Vendor (vmware).
CVSS VectorVendor: vmware
DescriptionCVE.org
The Spring Boot language server logs the raw value of the https_proxy/HTTPS_PROXY/http_proxy/HTTP_PROXY environment variable at INFO level whenever it creates an outbound HTTP client and no explicit http.proxy workspace setting is configured. Corporate proxy URLs frequently embed Basic-auth credentials in the form http://user:pass@proxy:8080, and the language server writes this value to its log file without any redaction. Since language server log files are often attached to bug reports or are readable by other local users/processes, this can result in disclosure of proxy credentials. Affected Spring Products and Versions: Spring Tools for Eclipse: 5.2.0 and earlier Spring Tools for VSCode / Cursor / Theia: 2.2.0 and earlier
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Information disclosure occurs when an application unintentionally exposes sensitive data that aids attackers in reconnaissance or directly compromises security.
Access control bypass in SpringBlade v4.5.0 importUser function allows low-privileged users to import sensitive user dat
Spring Cloud Config Server contains a path traversal vulnerability when using the native file system backend, allowing u
Spring Framework MVC applications can be vulnerable to a “Path Traversal Vulnerability” when deployed on a non-compliant
SpringBlade v4.5.0 has an access control flaw in authRoutes allowing low-privileged users to escalate to admin through t
Remote code execution in Spring for GraphQL versions 1.3.0-1.3.8, 1.4.0-1.4.5, and 2.0.0-2.0.3 allows unauthenticated at
Server-Sent Events (SSE) stream corruption affects Spring MVC applications built on the functional web framework across
Spring AI versions 1.0.0 through 1.0.4 and 1.1.0 through 1.1.3 allow unauthenticated remote code execution through Sprin
Unsafe JSON deserialization in Spring Framework's JMS message converters (MappingJackson2MessageConverter and JacksonJso
Spring Security Aspects may not correctly locate method security annotations on private methods. Rated critical severity
Safety-guard bypass in the Spring Framework SpEL engine (VMware/Broadcom) lets attacker-controlled expression input esca
Directory traversal in Spring Cloud Config server module allows remote unauthenticated attackers to read arbitrary files
Spring Security fails to properly write HTTP response headers in servlet applications across multiple versions (5.7.0-5.
EUVD-2026-50979
GHSA-9j65-vr9q-w452
NVD
CVE.org
EPSS
Vulnrichment
VulDB
Exploit-DB