Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessReach exposed RAGFlow trace_mindmap endpoint
-
ExploitSupply another tenant/object identifier
-
ExecutionServer performs data-access without ownership check
-
ImpactRetrieve cross-tenant data
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A01:2025 - Broken Access Control
ASVS 5.0
V8 - Authorization Access control enforcement
MITRE ATT&CK
1 technique across 1 tactic: Defense Evasion The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | The specific precondition is that the externally reachable trace_mindmap path accepts a caller-supplied object or tenant identifier and performs its data-access operation without enforcing any owner, tenant, workspace, or membership binding; an attacker simply substitutes another tenant's/object's identifier to read data they should not access. Exploitation is limited by the need for network reachability to the RAGFlow 0.24.0 instance and, realistically for a multi-tenant IDOR, likely a valid low-privileged account to reach the endpoint - though the supplied vector claims PR:N (unauthenticated), which conflicts with typical multi-tenant design and should be confirmed. The flaw is tied to the specific vulnerable version (0.24.0) and the trace_mindmap feature being deployed/enabled. |
| Impact | Signals conflict sharply and must be weighed together. The input CVSS is 9.8 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H), which paints a worst-case remote-unauthenticated full-compromise picture. However, the description characterizes a CWE-284 incorrect-access-control / IDOR flaw in the trace_mindmap path - a missing owner/tenant/membership binding on a data-access operation - which is fundamentally a data-exposure issue, so the I:H and A:H impacts in the supplied vector appear overstated for what is essentially unauthorized read access. EPSS is only 0.14% (3rd percentile), indicating very low modeled exploitation probability, and the CVE is NOT on CISA KEV, so there is no evidence of active exploitation. A public gist (Ro1ME) is referenced and likely contains proof-of-concept/write-up details, meaning exploit information is probably public even though it is not weaponized or actively used. No SSVC decision data (exploitation/automatable/technical-impact) was provided in the input, so that framework cannot be applied. CONFLICT TO VERIFY: the vector asserts PR:N (unauthenticated), yet a cross-tenant IDOR in a multi-tenant application typically requires a valid low-privileged session to reach the endpoint before pivoting to other tenants' identifiers - authentication requirements should be confirmed with the vendor. Net assessment: this is a genuine, easy-to-exploit authorization flaw if an instance is exposed, but given the niche product footprint, very low EPSS, and absence of KEV, it is a moderate real-world priority rather than an emergency - treat it as a prompt-patch item, not a drop-everything incident. |
| Exploit ScenarioDerived | An attacker able to reach a RAGFlow 0.24.0 instance sends a request to the trace_mindmap path while supplying another tenant's or object's identifier; because the endpoint performs the data-access operation without verifying that the caller owns or belongs to that tenant/workspace, the server returns data belonging to other tenants. A public gist referencing the issue suggests proof-of-concept details are available, lowering the effort required to reproduce it. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 05, 2026
Incorrect access control in infiniflow RAGFlow 0.24.0 lets remote attackers reach the trace_mindmap path, supply a caller-chosen object or tenant identifier, and have that identifier flow into a data-access operation that enforces no owner, tenant, workspace, or membership binding - effectively an IDOR that exposes another tenant's RAG data. The vendor-agnostic description and the NVD-listed CVSS 3.1 score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) conflict with our independent assessment, which carries the same network reachability and low complexity but scores it PR:L with confidentiality-only impact (C:H/I:N/A:N), and that discrepancy should be resolved before treating the flaw as unauthenticated pre-auth data compromise. CISA KEV does not list this CVE and EPSS is only 0.14% (3rd percentile), so confirmed active exploitation is absent; no public exploit code is identified at time of analysis, although the referenced GitHub gist is publicly accessible technical material that should be reviewed as a likely proof-of-concept.
RemediationAI
No vendor-released patch is identified in the available data, so the primary action is to upgrade RAGFlow past 0.24.0 as soon as the upstream project publishes a fixed release, verify the fix specifically addresses the trace_mindmap authorization check, and treat any pre-0.24.x-to-0.24.0 deployments as affected until proven otherwise; reference the NVD entry (https://nvd.nist.gov/vuln/detail/CVE-2026-51893) and ENISA EUVD-2026-91117, and review the public gist at https://gist.github.com/Ro1ME/79163b36f35460b266bc80120caf9da8 for the reproduction details that will confirm which object/tenant identifiers are accepted. Until a fix is available, the most effective compensating controls are: disable or remove the trace_mindmap route entirely if mindmap tracing is not required (side effect: loss of that feature, but it eliminates the attack surface); if it must stay, place the endpoint behind a reverse proxy or API gateway rule that restricts it to trusted internal networks and authenticated sessions (side effect: legitimate remote users lose access and must use a VPN or internal route); and enforce tenant/ownership checks at the application or data-access layer, for example by scoping every trace_mindmap query to the authenticated principal's tenant/workspace and rejecting identifiers that do not resolve within it. Note the trade-off on each: disabling the route breaks a product feature, network restrictions degrade the multi-tenant SaaS use case, and any application-layer patch is a local fork that must be re-applied on every upgrade. Because the supplied PR:N vector conflicts with our PR:L assessment, do not rely on perimeter authentication alone as the control - assume an attacker with a valid low-privileged account can substitute another tenant's identifier, and additionally log and alert on trace_mindmap requests whose identifiers do not match the caller's session tenant.
Affected VersionsENISA
Affects your stack
Find out whether the stacks and servers you run are on an affected version.
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
The flaw is an instance of CWE-284 (Improper Access Control) manifesting as a classic Insecure Direct Object Reference in RAGFlow's trace_mindmap functionality. RAGFlow is a multi-tenant retrieval-augmented-generation engine in which documents, knowledge bases, chat traces, and mindmap/trace artifacts are logically partitioned by owner, tenant, workspace, and membership records in a backing datastore. In a correct implementation the trace_mindmap handler would resolve the incoming identifier and then join or check against the authenticated principal's tenant/membership before issuing the read; here the caller-controlled object or tenant identifier is passed straight into the data-access layer with no such binding predicate, so the query succeeds for any identifier the caller guesses or enumerates. The root cause is therefore an authorization gap placed at the wrong layer (the endpoint trusts the identifier as proof of entitlement) rather than a cryptographic, injection, or parsing defect. The CPE supplied with the intelligence feed is unmapped (cpe:2.3:a:n/a:n/a), so the only reliable scope anchor is the affected-version string RAGFlow 0.24.0, corroborated by the ENISA EUVD entry EUVD-2026-91117. Note the tension in the supplied vectors: the NVD-style vector claims no privileges (PR:N) and full integrity/availability loss, whereas our independent assessment holds that a multi-tenant IDOR of this shape realistically requires an authenticated low-privileged session to reach the endpoint and yields unauthorized read (C:H/I:N/A:N) rather than write or denial of service; the trace_mindmap endpoint also has to be deployed and network-reachable for any exploitation to occur.
Know when Ragflow 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:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Network-reachable IDOR with no complexity or user interaction; PR:L because reaching a multi-tenant endpoint typically needs a session, and impact is unauthorized data read (C:H) with no integrity/availability effect.
Primary rating from Vendor (mitre).
CVSS VectorVendor: mitre
Lifecycle Timeline
3DescriptionCVE.org
infiniflow ragflow 0.24.0 is vulnerable to Incorrect Access Control via trace_mindmap. An externally reachable path accepts a caller-selected object or tenant identifier and reaches a data-access operation without a visible owner, tenant, workspace, or membership binding on that object.
Executive ViewAI
Organizations running RAGFlow 0.24.0 face a critical access-control weakness that could allow one tenant to access another tenant's RAG data via the trace_mindmap path, undermining the isolation assumptions of multi-tenant AI knowledge bases. No vendor-released patch identified at time of analysis, and while CISA KEV does not list this CVE and no public exploit identified at time of analysis, the potential for cross-tenant exposure of proprietary or regulated content warrants immediate containment. Leadership should prioritize identifying affected RAGFlow 0.24.0 deployments and treating their RAG data as at risk until the authentication dispute and vendor remediation are resolved.
No vendor-released patch identified at time of analysis. Interim mitigations include restricting network access to RAGFlow 0.24.0, especially the trace_mindmap path, to trusted users and networks; disabling or blocking trace_mindmap if not required; enforcing tenant, workspace, and membership authorization at an upstream reverse proxy or API gateway; deploying WAF rules to detect identifier manipulation; enabling and monitoring audit logs for cross-tenant data access; applying least privilege and rotating credentials or API keys if suspicious activity is found; and applying the vendor patch as soon as it is released.
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
Authentication bypass attacks exploit flaws in the verification mechanisms that control access to systems and applications.
A vulnerability in infiniflow/ragflow versions v0.12.0 allows for remote code execution. Rated critical severity (CVSS 9
Path traversal vulnerability in RAGFlow RAG engine version 0.23.1 allows unauthenticated attackers to read arbitrary fil
In infiniflow/ragflow versions 0.12.0, the `web_crawl` function in `document_app.py` contains multiple vulnerabilities.
RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine. Rated high severity (CVSS 8.9), this vulnerabilit
RAGFlow versions prior to 0.22.0 generate API keys and beta assistant/agent share tokens with a predictable scheme, lett
The `add_llm` function in `llm_app.py` in infiniflow/ragflow version 0.11.0 contains a remote code execution (RCE) vulne
Remote code execution in RAGFlow versions prior to 0.23.0 allows a low-privileged authenticated user (a normal login acc
RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding. Rated high s
A Server-Side Request Forgery (SSRF) vulnerability exists in infiniflow/ragflow version 0.12.0. Rated high severity (CVS
RAGFlow 0.13.0 suffers from improper access control in document-hooks.ts, allowing unauthorized access to user documents
A vulnerability in infiniflow/ragflow version RAGFlow-0.13.0 allows for partial account takeover via insecure data query
Server-side template injection in RAGFlow 0.24.0 and earlier allows any authenticated user to execute arbitrary operatin
EUVD-2026-91117
GHSA-p84p-8f64-q95f
NVD
CVE.org
EPSS
Vulnrichment
Exploit-DB