Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessAuthenticate as low-privilege user
-
DeliveryOpen/edit chatflow with CSV Agent node
-
ExploitInject obfuscated payload in customReadCSV
-
ExecutionBypass regex denylist validator
-
PersistExecute Python in unsandboxed pyodide
-
ImpactFull host compromise
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A05:2025 - Injection
ASVS 5.0
V1 - Encoding and Sanitization Code injection prevention, sandbox exec
MITRE ATT&CK
2 techniques across 2 tactics: Execution, Initial Access The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | Exploitation requires an authenticated Flowise account with permission to create or modify a chatflow that uses the CSV Agent node, and the ability to set that node's customReadCSV parameter (CVSS PR:L confirms low-privilege authentication is needed; UI:N means no victim interaction is required once the attacker triggers the node). The specific limiting factor is the AC:H obfuscation requirement - the attacker must craft input that evades the static regex blocklist rather than submit plain Python - and access to the CSV Agent node feature itself; instances that do not expose chatflow-editing to untrusted users, or that are network-restricted, materially reduce reachability. The impact hinges on the deployment shipping the unsandboxed pyodide runtime, which the advisory states is the default. |
| Impact | The signals are mostly consistent and point to a genuine high-severity issue for exposed instances: the CVSS 4.0 vector (AV:N/AC:H/PR:L/UI:N with high vulnerable- and subsequent-system impact, base 9.0) reflects network reach, low-privilege authentication, no user interaction, and a scope-crossing impact into the host. The one moderating factor is AC:H, which aligns with the description's requirement to craft an obfuscation payload that defeats the regex denylist - meaning exploitation is reliable for a knowledgeable attacker but not trivially point-and-click. Because the CSV Agent node is a normal product feature, a mid-sized operator running Flowise for internal AI tooling should treat any authenticated user (or any compromised low-privilege account) as a potential path to full server compromise, and prioritize the 3.1.3 upgrade regardless of perimeter posture. There is no KEV listing and no EPSS score supplied, and no public POC is referenced, so real-world exploitation should currently be judged as plausible-but-unproven rather than active; the disclosure by VulnCheck plus a published GitHub Security Advisory raises the likelihood that working exploits appear, so patching should not be deferred on the basis of missing EPSS/KEV data alone. |
| Exploit ScenarioDerived | An attacker with a low-privilege Flowise account (or a phished/compromised builder credential) opens or creates a chatflow, adds or edits a CSV Agent node, and supplies an obfuscated Python payload in customReadCSV that slips past the regex denylist. When the node runs, the code executes in the unsandboxed pyodide environment and gains full access to the underlying host, enabling data theft, lateral movement, or persistence. No public POC is referenced, but the AC:H obfuscation technique is well understood for denylist bypasses. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Aug 13, 2026
Authenticated arbitrary Python code execution in FlowiseAI Flowise before 3.1.3 arises from the CSV Agent node's customReadCSV parameter, whose regex-based validator can be defeated with obfuscation to run attacker-supplied code inside an unsandboxed pyodide runtime with full host access. Any user able to build or edit a chatflow can escalate from application-level access to code execution on the server. No public exploit has been identified at time of analysis, and the issue is not listed in CISA KEV; a vendor patch exists in version 3.1.3.
RemediationAI
Vendor-released patch: 3.1.3 - upgrade Flowise to 3.1.3 or later, which is the primary and confirmed fix per GHSA-4878-cqgq-j53v (https://github.com/FlowiseAI/Flowise/security/advisories/GHSA-4878-cqgq-j53v) and the VulnCheck advisory (https://www.vulncheck.com/advisories/flowise-before-code-injection-via-csv-agent-customreadcsv). If immediate patching is not possible, reduce exposure by restricting who can create or edit chatflows - tighten authentication and remove low-privilege build/edit rights so untrusted accounts cannot reach the CSV Agent node (trade-off: limits legitimate self-service workflow authoring), and avoid using or disable the CSV Agent node where feasible. Additionally place the Flowise instance behind network access controls so its UI/API is not internet-reachable and only trusted users can authenticate (trade-off: may impede remote/collaborative use), and monitor the pyodide execution paths and node configuration changes for anomalous Python payloads. Because the underlying control is a regex denylist, do not rely on input filtering as a durable mitigation - the fixed version is the only reliable control.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Flowise is an open-source low-code platform for visually building LLM/agent workflows, where individual 'nodes' expose configurable parameters. The affected CSV Agent node accepts a customReadCSV parameter intended to hold Python parsing logic that executes in a bundled pyodide (WebAssembly CPython) interpreter. The root cause is CWE-94 (Improper Control of Generation of Code / code injection): input is gated by a static regex denylist rather than a real sandbox or allowlist, and denylist filtering is inherently bypassable via encoding/obfuscation (e.g., string concatenation, getattr/__import__ tricks, character escapes). Because the pyodide environment is described as unsandboxed with full system access, escaping the language-level filter yields host-level code execution rather than confined interpreter execution. Affected CPE is cpe:2.3:a:flowiseai:flowise across all versions prior to the fix.
Know when Flowise 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:H/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H
Network-reachable but requires a low-privilege account (PR:L) and crafting an obfuscated denylist bypass (AC:H); code escapes the interpreter to the host, so scope changes (S:C) with full C/I/A impact.
Primary rating from Vendor (VulnCheck).
CVSS VectorVendor: VulnCheck
Lifecycle Timeline
2DescriptionCVE.org
Flowise before 3.1.3 contains a code injection vulnerability in the CSV Agent node's customReadCSV parameter that allows authenticated attackers to execute arbitrary Python code. The validator uses a static regex blocklist that can be bypassed through obfuscation techniques, enabling attackers to execute code in the unsandboxed pyodide environment with full system access.
Articles & Coverage 1
Executive ViewAI
FlowiseAI Flowise before version 3.1.3 permits authenticated chatflow editors to execute arbitrary server-side code through the CSV Agent node, converting application-level access into complete infrastructure control. This affects any organization using Flowise for workflow automation and requires immediate patching to prevent insider privilege escalation.
Apply vendor patch.
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
Code injection occurs when an application accepts user input and passes it directly into a language interpreter or evaluator without proper sanitization.
Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.
Flowise version 3.0.5 contains a remote code execution vulnerability in the CustomMCP node. The mcpServerConfig paramete
Flowise versions before 3.0.1 allow unauthenticated access to the Custom MCPs feature, which is designed to execute OS c
FlowiseAI Flowise version 2.2.6 contains an arbitrary file upload vulnerability in the /api/v1/attachments endpoint. Una
Arbitrary file read and write in Flowise versions prior to 3.0.8 allows an authenticated attacker with access to the flo
Remote code execution in Flowise before 3.1.2 allows any authenticated user (or API caller with chatflow view/update per
Flowise is a drag & drop user interface to build a customized large language model flow. Rated critical severity (CVSS 9
Authenticated users can upload arbitrary files to Flowise 3.0.7 due to missing validation, enabling persistent storage o
Unrestricted file upload in Flowise LLM workflow builder before 3.0.13 via /api/v1/attachments endpoint allows unauthent
Missing authentication on NVD data endpoint in Flowise before 3.0.13 allows unauthenticated access to internal vulnerabi
Authenticated remote code execution in FlowiseAI Flowise (v3.0.1 up to but not including 3.0.8, and later versions when
Remote code execution in Flowise before 3.1.3 arises because its Pyodide Python-code validator (validatePythonCodeForDat
Remote code execution in FlowiseAI Flowise before 3.1.3 lets an attacker execute arbitrary OS commands as the Flowise pr
EUVD-2026-57802
GHSA-xcrh-wphv-74q4
NVD
CVE.org
EPSS
Vulnrichment
VulDB
Exploit-DB