Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessOpen unauthenticated WebSocket/SSE connection
-
DeliverySend mount frame with crafted view path
-
ExploitServer calls __import__ before auth/subclass check
-
ExecutionModule top-level code executes
-
ImpactDoS, import-time RCE-by-proxy, or enumeration oracle
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A05:2025 - Injection
ASVS 5.0
V15 - Secure Coding and Architecture Reflection safety, class allowlisting
|
| Exploitation Conditions | Reachable against the DEFAULT configuration of djust < 1.0.7: the LIVEVIEW_ALLOWED_MODULES setting is unset by default, and the fail-open |
| Impact | No CVSS score or vector is provided by NVD or the vendor, so severity signals must be reasoned from the description. The GitHub Security Advisory (GHSA-7prp-2623-8g45) explicitly states an UNAUTHENTICATED WebSocket/SSE client can trigger the flaw, that the WS handshake requires no auth, and that it was reproduced end-to-end - placing this at network attack vector, low complexity, no privileges, no user interaction. CWE-470 (unsafe reflection) fits: a client-supplied dotted path is passed to |
| Exploit ScenarioDerived | An attacker opens an unauthenticated WebSocket (or SSE) connection to a djust-powered Django app running the default configuration (LIVEVIEW_ALLOWED_MODULES unset) and sends a |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Sep 16, 2026
Unauthenticated arbitrary Python module import in djust before 1.0.7 lets any client that can reach the WebSocket or SSE mount endpoint force the server to import a module chosen by a client-supplied dotted path, executing that module's import-time side effects before any LiveView subclass check or per-view authentication occurs. The LIVEVIEW_ALLOWED_MODULES allowlist that should constrain this lookup is fail-open (the check is skipped when the setting is unset, which is the framework default) and uses loose startswith matching, so the flaw is reachable in a default configuration; the assessed vector is CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H. Impact is bounded by what is already installed and importable on the host: guaranteed outcomes are denial of service (import bombs, expensive dependency trees) and a module/class enumeration oracle exposed through distinct error strings, escalating to RCE-by-proxy where a suitably side-effectful module exists. No public exploit has been identified at time of analysis - the maintainer retains the reproducer privately - and there is no CISA KEV listing.
RemediationAI
Upgrade to djust 1.0.7, the vendor-released patch, which adds the fail-closed djust._view_resolution.is_view_import_allowed gate ahead of every __import__ sink; no backport to earlier branches is indicated in the advisory. If an immediate upgrade is not feasible, set LIVEVIEW_ALLOWED_MODULES to the narrow list of modules that actually contain your mountable LiveView classes - this blocks the most obvious enumeration but is mitigation only, not a complete fix, because pre-patch matching is startswith-based and the import still runs before the LiveView subclass check, leaving prefix-similar and lazily importable modules reachable. Additional compensating controls carry trade-offs: restrict network access to the djust WebSocket and SSE endpoints via reverse-proxy allowlists, VPN, or an authenticated gateway to eliminate the unauthenticated attack surface, at the cost of breaking legitimate external clients; or run the process with a minimal set of importable modules and no credentials in its environment to limit the blast radius of import-time side effects, though this does not prevent denial of service from import bombs and heavy dependency trees. Where a proxy or gateway is available, rate-limiting mount frames and capping concurrent connections reduces the DoS vector but does not address the arbitrary-import primitive. Verify remediation by confirming the installed version is 1.0.7 or later (for example, pip show djust) and by re-sending an unauthenticated mount frame referencing an out-of-scope module path, which should now be rejected before import. References: https://github.com/djust-org/djust/security/advisories/GHSA-7prp-2623-8g45 and https://github.com/djust-org/djust/releases/tag/v1.0.7.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in pypi
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
djust is a Django-oriented LiveView framework whose live transport keeps server-rendered views synchronized over WebSocket and Server-Sent Events. The root cause is CWE-470 (Use of Externally-Controlled Input to Select Classes or Code): handle_mount in python/djust/websocket.py, ViewRuntime.dispatch_mount / _instantiate_view in python/djust/runtime.py (the SSE and url_change path), and the SSE mount in python/djust/sse.py all take the view identifier from the client frame and pass the dotted module portion straight into Python's __import__(module_path, ...). Because Python executes a module's top-level code on first import, the import call itself is the dangerous primitive, and in the vulnerable versions it occurs before the framework verifies that the resolved object is a LiveView subclass and before any per-view authentication. The setting intended to gate this, LIVEVIEW_ALLOWED_MODULES, is fail-open (if allowed_modules:), so an unset value - the framework default - bypasses enforcement entirely; when configured, it matches with startswith rather than on module-segment boundaries, allowing prefix-confusable module names through. Version 1.0.7 replaces this with a fail-closed resolution gate, djust._view_resolution.is_view_import_allowed, invoked before __import__ at all three sinks with an additional defensive check inside _instantiate_view: a client view path resolves only if its module is already present in sys.modules (so resolving runs no new code, and views loaded by the URLconf at startup continue to work with zero configuration) or if it matches LIVEVIEW_ALLOWED_MODULES on a module-segment boundary as an explicit opt-in for lazily imported views.
Know when Python 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
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:H/VA:L/SC:N/SI:N/SA:N
Unauthenticated network WebSocket/SSE with default fail-open config gives AV:N/AC:L/PR:N/UI:N; unconditional DoS (A:H) plus enumeration oracle and conditional import-time code execution justify high C/I, though full RCE depends on a side-effectful module.
Primary rating from Vendor (https://github.com/djust-org/djust).
CVSS VectorVendor: https://github.com/djust-org/djust
Lifecycle Timeline
2DescriptionCVE.org
Impact
The djust live transport resolves the LiveView to mount from a client-supplied dotted path by calling __import__(module_path, ...). The module is imported - running its top-level code (import side effects) - *before* the framework checks that the resolved object is a LiveView subclass and *before* any per-view authentication. The LIVEVIEW_ALLOWED_MODULES allowlist that should contain this is fail-open (if allowed_modules: - skipped when the setting is unset, the framework default) and uses loose startswith matching.
An unauthenticated WebSocket client (the WS handshake does not require auth; per-view auth runs only after import + instantiate) can therefore send a mount / live_redirect_mount / url_change frame (or an SSE mount) with view = "<any.importable.module>.AnyName" and cause the server to import - and execute the top-level code of - any importable Python module by name.
Consequences: server-side execution of arbitrary importable modules' import-time side effects by an unauthenticated client (effectively RCE-by-proxy on any host that has a side-effectful importable module), denial of service (import bombs / expensive dependency trees), and a module/class enumeration oracle via distinct error strings.
Reproduced end-to-end: an unauthenticated WebsocketCommunicator mount frame with the allowlist unset imported and executed a sentinel non-LiveView module before the "not a LiveView subclass" rejection.
Affected code
python/djust/websocket.pyhandle_mount(__import__of the clientview)python/djust/runtime.pyViewRuntime.dispatch_mount/_instantiate_view(SSE +url_changepath)python/djust/sse.pySSE mount
Threat-model entry T4 (docs/audits/websocket-auth-2026-06.md) previously noted the default-open allowlist but understated the impact as mere LiveView-class probing; the real primitive is arbitrary-module import + top-level code execution, independent of whether the target is a LiveView.
Patches
Fixed by a fail-closed resolution gate (djust._view_resolution.is_view_import_allowed): a client view path resolves only if (a) its module is already loaded (sys.modules - so resolving runs no new code; URL-routed views loaded by URLconf at startup keep working with zero config) or (b) it matches LIVEVIEW_ALLOWED_MODULES on a module-segment boundary (explicit opt-in for lazily-imported views). The gate runs before __import__ at all three sinks (+ defense-in-depth inside _instantiate_view).
Workarounds
Set LIVEVIEW_ALLOWED_MODULES to the narrow list of modules that contain your mountable LiveView classes. (Note: pre-patch the allowlist is startswith-matched and the import still precedes the subclass check, so this is mitigation, not a complete fix.)
References
Reproducer + finding writeup retained privately by the maintainer.
Technical POC Analysis
CWE-470 (Use of Externally-Controlled Input to Select Classes or Code, "Unsafe Reflection").
Full technical POC analysis - root cause, step-by-step exploitation and patch breakdown - is part of the Pro and Business plans.
Latest POC
1 PoC candidate from github (unverified)
Sign in to open the public PoC links collected for this CVE.
Executive ViewAI
The Python LiveView framework djust, in all versions before 1.0.7, allows any unauthenticated client that can reach an application's WebSocket or SSE mount endpoint to make the server import a module of the attacker's choosing before any authentication or view-level check runs. Because the intended allowlist safeguard is inactive in the framework's default configuration, exposed deployments can be disrupted, probed for installed components, or - where certain side-effectful libraries are present - driven toward code execution. Leadership should care because exploitation requires no credentials and no user interaction, so every internet-reachable djust deployment is in scope today, even where no public exploit code has been identified at time of analysis.
No vendor-released patch identified at time of analysis. Interim mitigations: explicitly configure LIVEVIEW_ALLOWED_MODULES as a strict, exact-match allowlist (the default leaves this check skipped and the existing matching is prefix-based and too permissive); restrict network reachability of the WebSocket and SSE mount endpoints to trusted clients; disable LiveView mounting where unused; minimize importable packages on the host to limit denial-of-service and proxy-execution exposure; and monitor server logs for distinct import-error strings indicating module/class enumeration.
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
Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.
Denial of Service attacks render applications or systems unavailable by overwhelming resources or triggering failure conditions.
Wazuh SIEM platform versions 4.4.0 through 4.9.0 contain an unsafe deserialization vulnerability in the DistributedAPI t
BentoML version 1.4.2 and earlier contains an unauthenticated remote code execution vulnerability through insecure deser
pgAdmin 4 contains critical remote code execution vulnerabilities in the Query Tool download and Cloud Deployment endpoi
The renderLocalView function in render/views.py in graphite-web in Graphite 0.9.5 through 0.9.10 uses the pickle Python
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Rated critica
OpenSSL before 0.9.8za, 1.0.0 before 1.0.0m, and 1.0.1 before 1.0.1h does not properly restrict processing of ChangeCiph
pyLoad download manager version prior to 0.5.0b3.dev77 exposes the Flask SECRET_KEY through an unauthenticated endpoint.
Langflow (a visual LLM pipeline builder) contains a critical unauthenticated code execution vulnerability (CVE-2026-3301
In Mercurial before 4.1.3, "hg serve --stdio" allows remote authenticated users to launch the Python debugger, and conse
Unauthenticated remote code execution affects Kestra OSS (the open-source event-driven orchestration platform) prior to
Unauthenticated remote code execution in Marimo ≤0.20.4 allows attackers to execute arbitrary system commands via the `/
pyLoad is the free and open-source Download Manager written in pure Python. Rated medium severity (CVSS 5.3), this vulne
EUVD-2026-81307
GHSA-7prp-2623-8g45
NVD
CVE.org
EPSS
Vulnrichment
VulDB
Exploit-DB