Severity by source
CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Primary rating from Vendor (https://github.com/FlowiseAI/Flowise) · only source for this CVE.
CVSS VectorVendor: https://github.com/FlowiseAI/Flowise
CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Lifecycle Timeline
7DescriptionCVE.org
Summary
Type: Mass assignment via Object.assign(entity, body) -> client-controlled workspaceId (and on create, id) overwritten on the Dataset entity -> cross-workspace data takeover and IDOR. File: packages/server/src/services/dataset/index.ts Root cause: The Dataset controller/service constructs a new Dataset() and copies the request body into it via Object.assign(...) without an explicit field allowlist. The request body therefore can include workspaceId, id, createdDate, updatedDate. The server only rebinds *some* of these after the assign (e.g. on create, it overwrites workspaceId but not id; on update, it overwrites id but not workspaceId). The remaining client-controlled values land directly on the persisted row, breaking workspace isolation. Same root pattern as the dataset entity's sibling controllers and as DocumentStore before it was patched in commit 840d2ae.
Affected Code
File: packages/server/src/services/dataset/index.ts
// create (line 203) and update (line 226)
Object.assign(newDataset, body) // <-- BUG: body.id, body.workspaceId acceptedWhy it's wrong: Object.assign(target, source) copies every own enumerable property of source onto target. The TypeORM/SQL persistence layer below it does not strip ownership-bearing columns, so workspaceId set in the request body lands as the new workspaceId of the persisted row. The DocumentStore patch (commit 840d2ae) demonstrated the intended fix shape (explicit field-by-field allowlist) but it has not been applied to this entity.
Exploit Chain
- Attacker is an authenticated member of workspace A. They have a session cookie / JWT for the Flowise web UI. State at this point: attacker can read and write entities scoped to workspace A.
- Attacker creates a dataset in workspace A via the documented API (or reuses an existing one they own). They note its entity
id. - Attacker issues a
PUT /api/v1/datasets/<id>(or equivalent endpoint) with a JSON body that includes"workspaceId": "<workspace-B-id>"(an arbitrary other workspace's UUID). State at this point: the request reaches the controller as a workspace-A authenticated request. - The controller calls
Object.assign(updateEntity, body). The body'sworkspaceIdoverwrites the entity'sworkspaceIdfield. The persistence layer commits the row. - Final state: the dataset row is now owned by workspace B. Workspace B members can see it, modify it, and use it. Workspace A loses access (it no longer satisfies their workspace filter). The original creator's workspace audit shows nothing because the operation looked like a normal update.
Security Impact
Severity: High. Cross-workspace boundary violation by any authenticated workspace member. Attacker capability: Any authenticated user with permission to update a dataset can move it to any workspace whose UUID they can guess or enumerate (workspace UUIDs are exposed in many API responses, so enumeration is trivial). Datasets hold training / evaluation data scoped to a workspace. Moving a Dataset across workspaces via workspaceId overwrite exposes the dataset (rows, schema, references) to the destination workspace. Preconditions: Authenticated session with edit permission for the source dataset. No second factor required. Workspace UUIDs are exposed via the /api/v1/workspaces listing or via any cross-referenced object's workspaceId field, so target enumeration is trivial. Differential: PoC-verified by source inspection of the original GHSA-q4pr-4r26-c69r. Patched build (with the suggested fix below) refuses the workspaceId field; vulnerable build accepts it and persists it.
Suggested Fix
Already fixed in PR https://github.com/FlowiseAI/Flowise/pull/6051 (allowlist pattern applied).
// Allowlist pattern (matches commit 840d2ae for DocumentStore):
const updatedDataset = new Dataset()
if (body.<allowed_field_1> !== undefined) updatedDataset.<allowed_field_1> = body.<allowed_field_1>
if (body.<allowed_field_2> !== undefined) updatedDataset.<allowed_field_2> = body.<allowed_field_2>
// ...whitelist only the documented fields. Never copy id, workspaceId, createdDate, updatedDate from the client.Regression tests should assert that a request body containing workspaceId, id, createdDate, or updatedDate is rejected (or at minimum: does not change those columns on the persisted row) for both create and update paths.
AnalysisAI
Cross-workspace dataset takeover in FlowiseAI Flowise (<=3.1.1) allows any authenticated workspace member to reassign Dataset entities to an arbitrary workspace by submitting a crafted workspaceId field in create/update request bodies. The flaw stems from an unfiltered Object.assign(entity, body) call in packages/server/src/services/dataset/index.ts, breaking tenant isolation in this multi-workspace LLM orchestration platform. No public exploit identified at time of analysis, though the GHSA-5h9v-837x-m97r advisory includes step-by-step exploitation details and the vendor confirms PoC verification via source inspection.
Technical ContextAI
Flowise is a popular open-source low-code LLM application builder (npm package 'flowise') that organizes resources into workspaces for multi-tenant isolation. The root cause is CWE-915 (Improperly Controlled Modification of Dynamically-Determined Object Attributes), commonly known as mass assignment: the Dataset controller constructs a new TypeORM entity and uses Object.assign(newDataset, body) at lines 203 (create) and 226 (update) without an explicit field allowlist. Because Object.assign copies every own enumerable property, client-supplied keys like workspaceId, id, createdDate, and updatedDate are persisted directly to the database row. The create path overwrites workspaceId post-assign but leaves id client-controlled; the update path overwrites id but leaves workspaceId client-controlled - the inverse gap on each path. The same anti-pattern was previously fixed in DocumentStore (commit 840d2ae) using the allowlist shape now applied here.
RemediationAI
Vendor-released patch: upgrade flowise to 3.1.2 or later, which applies the allowlist pattern from commit 49a2259bf2a6b4f3d4b50813cb5161cee0d40040 (PR https://github.com/FlowiseAI/Flowise/pull/6051) and mirrors the DocumentStore fix at commit 840d2ae. Operators who cannot upgrade immediately should restrict dataset write permissions to highly trusted users only and add an API gateway or reverse-proxy rule that strips workspaceId, id, createdDate, and updatedDate from JSON bodies on POST/PUT to /api/v1/datasets - note that body-rewriting at the proxy can break legitimate client schemas if they depend on echoing these fields, and stripping id on PUT will not work if your routing requires it. As a second compensating control, audit existing Dataset rows for unexpected workspaceId changes by joining against application/audit logs, since the original advisory notes the mutation appears as a normal update with no special audit signal. Reference the GitHub advisory https://github.com/FlowiseAI/Flowise/security/advisories/GHSA-5h9v-837x-m97r for authoritative guidance.
Share
External POC / Exploit Code
Leaving vuln.today
EUVD-2026-35114
GHSA-5h9v-837x-m97r