Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessAcquire page-author credentials
-
DeliveryCraft sandboxed Twig payload with grav.offsetGet('config')
-
ExploitApply serialization filter bypassing GravSecurityPolicy
-
ExecutionRender page to receive raw config output
-
ImpactExtract SMTP credentials, API keys, and database passwords from response
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A01:2025 - Broken Access Control
ASVS 5.0
V14 - Data Protection Data exposure prevention
MITRE ATT&CK
2 techniques across 2 tactics: Collection, Discovery 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 Grav user account with page-author permissions - a role that must be explicitly granted by an administrator and is not present by default on a fresh Grav installation with a single admin owner. The Twig sandbox must be active (it is the default sandboxing mode for page-author-level content in Grav). The vulnerability affects the default configuration of any Grav instance prior to 2.0.2 where at least one page-author account exists. Exploitation is strictly limited to confidentiality impact: no code execution, no file write, no availability impact. Attackers without an active page-author session cannot exploit this vulnerability remotely. |
| Impact | The CVSS 3.1 vector AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N scores 6.5, consistent with the description: the attack is network-reachable with no complexity barrier once an authenticated page-author account is held, with full confidentiality impact limited to data within the application. The PR:L prerequisite is the primary limiting factor - an attacker must already possess a page-author credential, which is a non-default, intentionally granted role. No EPSS score was included in the provided intelligence, so probabilistic exploitation likelihood cannot be quantified from this dataset; the absence should be noted rather than interpolated. The vulnerability is not listed in CISA KEV, and no standalone POC is confirmed beyond the public GitHub security advisory and commit diff, which together provide sufficient technical detail to reconstruct the attack. Real-world impact is significant for any multi-tenant or community-driven Grav deployment where untrusted contributors hold page-author permissions: credential exfiltration enables downstream account takeover against SMTP servers, external APIs, and databases referenced in plugin configuration - a meaningful blast radius despite the moderate CVSS score. Scope is unchanged (S:U), correctly reflecting that impact remains within the Grav application boundary even though harvested credentials may enable lateral movement outside it. |
| Exploit ScenarioDerived | An attacker with a Grav page-author account edits or creates a page containing sandboxed Twig content such as {{ grav.offsetGet('config')|json_encode }} or {{ grav.offsetGet('config')|print_r }}, then previews or publishes the page to trigger sandbox rendering. The serialization filter bypasses GravSecurityPolicy and outputs the full plugins.* configuration tree - including SMTP credentials and API keys - in the rendered page response, which the attacker reads directly. No POC is publicly confirmed, but the public commit diff and advisory provide sufficient detail to reconstruct this technique without additional research. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Aug 19, 2026
Configuration secret exposure in Grav (all versions prior to 2.0.2) allows authenticated page authors to exfiltrate sensitive plugin credentials - including SMTP passwords, API keys, and plugin database credentials - through a Twig sandbox bypass. The Grav container's offsetGet and __get methods were allow-listed, enabling page-authored Twig to obtain the raw Config object, while serialization filters (json_encode, print_r, yaml_encode, string) were not routed through GravSecurityPolicy::checkMethodAllowed, so the sandbox's member-access gate was never triggered. The vendor-released fix is confirmed in version 2.0.2; no public exploit code or CISA KEV listing is confirmed at time of analysis.
RemediationAI
Upgrade Grav to version 2.0.2 or later; this is the vendor-released patch confirmed by the GitHub release at https://github.com/getgrav/grav/releases/tag/2.0.2 and the commit at https://github.com/getgrav/grav/commit/7efe70574e4cf01a251318478c13673029e93eb6. If immediate upgrade is not feasible, the only effective compensating control is to audit and revoke page-author permissions from any untrusted or externally-facing user accounts, since the attack is gated entirely on that role; this has the trade-off of removing legitimate content-editing capability for those users. There is no documented configuration toggle that disables the vulnerable filter registrations without patching, because the serialization filters (json_encode, print_r, yaml_encode, string) are core Twig extension functionality. Rotating any credentials stored in Grav plugin configuration (SMTP passwords, API keys, database passwords) is advisable for any deployment that has granted page-author access to untrusted parties prior to patching.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 2 packages in composer, euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Grav is a PHP flat-file CMS using the Twig templating engine with a sandboxed rendering mode (GravSecurityPolicy) intended to restrict what page-author-level users can access in templates. The root cause (CWE-200: Exposure of Sensitive Information to an Unauthorized Actor) is twofold. First, the system/config/security.yaml allow-list for the Grav DI container included the offsetget and __get methods, which return raw service objects - including the real Config instance that holds all plugins.* configuration values. Second, serialization filters (print_r, json_encode, yaml_encode, string, vardump) were registered without sandbox guards, so they never called GravSecurityPolicy::checkMethodAllowed or checkPropertyAllowed when processing objects that had entered sandbox scope through the allow-listed container access. The fix (commit 7efe705) removes offsetget and __get from the method allow-list and wraps all dump/serialize filters with an assertSandboxDumpSafe guard that recursively validates objects against the active policy before allowing serialization. Affected CPE: cpe:2.3:a:getgrav:grav:*:*:*:*:*:*:*:* covering all versions prior to 2.0.2.
Know when Grav 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:L/UI:N/S:U/C:H/I:N/A:N
Network-reachable attack requiring only a page-author session (PR:L); no integrity or availability impact; full confidentiality loss of plugin credential secrets justifies C:H.
Primary rating from Vendor (GitHub_M).
CVSS VectorVendor: GitHub_M
Lifecycle Timeline
1DescriptionCVE.org
Grav is a file-based Web platform. Prior to 2.0.2, the Grav Twig content sandbox permits grav.offsetGet('config') to return the raw configuration object and permits json_encode, print_r, yaml_encode, and string filters to serialize that object without passing through GravSecurityPolicy::checkMethodAllowed. A user with page-author permissions can render sandboxed content that exposes plugins.* configuration secrets, including SMTP credentials, API keys, and plugin database credentials. This issue is fixed in version 2.0.2.
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.
Server-side template injection (SSTI) in Grav (getgrav/grav) prior to 1.8.0-beta.27 allows an authenticated attacker hol
Broken access control in Grav CMS (CWE-285) before 1.8.0-beta.27 lets an authenticated editor-level user escalate their
A Remote Code Execution (RCE) vulnerability in Grav CMS v1.7.48 allows an authenticated admin to upload a malicious plug
Grav is a file-based Web platform. Rated critical severity (CVSS 9.9), this vulnerability is remotely exploitable, low a
GravCMS 1.10.7 allows unauthenticated remote attackers to write arbitrary YAML configuration files, leading to full serv
Remote code execution in Grav CMS 2.0.4 (fixed in 2.0.7) lets an authenticated user holding the admin.pages or api.pages
Remote code execution in Grav CMS before 2.0.15 allows attackers holding page-edit or blueprint-config privileges to wri
Vertical privilege escalation in Grav CMS before 2.0.14 lets a delegated administrator holding only the admin.users perm
Stored cross-site scripting in Grav CMS before 2.0.15 lets authenticated editors smuggle JavaScript event handlers past
Server-side request forgery (SSRF) in Grav CMS versions prior to 1.7.49.5 allows remote authenticated attackers to make
Grav is a file-based Web platform. Prior to 1.8.0-beta.27, Grav CMS is vulnerable to a Server-Side Template Injection (S
Remote code execution in Grav CMS before 2.0.13 allows authenticated low-privilege users to plant and execute arbitrary
EUVD-2026-62535
GHSA-mc5q-6hpj-rp7j
NVD
CVE.org
EPSS
Vulnrichment
VulDB
Exploit-DB