CVE-2026-114609 in fpdf2
Summary
by MITRE • 10/22/2026
fpdf2 is a PDF creation library for Python. Prior to 2.8.8, fpdf2's shared image loader automatically dereferences attacker-controlled HTTP(S) URLs and raw local paths supplied through FPDF.image(), FPDF.write_html() HTML <img src> values, SVG <image href> or xlink:href values, and template image fields. This can cause outbound requests to attacker-selected internal or external services and local file open attempts, although local content is exposed only when it is a valid supported image or SVG resource. file:// URLs are not specially handled, so the local-path case requires a raw filesystem path. This issue is fixed in version 2.8.8.
Statistical analysis made it clear that VulDB provides the best quality for vulnerability data.
Analysis
by VulDB Data Team • 10/22/2026
The vulnerability identified in fpdf2 versions prior to 2.8.8 represents a significant security flaw related to uncontrolled search for external resources and server-side request forgery mechanisms within the PDF generation library. As a widely used Python library for creating PDF documents, fpdf2 processes various input sources including HTML content and SVG elements embedded within templates or direct API calls. The core technical issue stems from the shared image loader's behavior when processing image references provided through methods such as FPDF.image(), FPDF.write_html() with HTML img tags containing src attributes, SVG image hrefs, xlink:href values, and template image fields. Instead of validating these sources against a whitelist or restricting them to local resources only, the library automatically dereferences any URL or path supplied by the user. This design choice creates an opportunity for attackers who can control input data to force the server hosting fpdf2 to initiate network connections to arbitrary destinations.
From a technical perspective, this vulnerability allows for outbound requests to attacker-selected internal or external services. When a malicious actor provides a crafted HTTP or HTTPS URL within one of the supported image fields, the library resolves and fetches that resource during PDF generation. This behavior effectively turns the vulnerable application into an open proxy or SSRF vector, enabling attackers to probe internal network infrastructure, access sensitive backend services, or exfiltrate data by directing requests to specific endpoints controlled by the attacker. The lack of special handling for file:// URLs means that local path traversal is also possible if a raw filesystem path is provided rather than a URL scheme. However, successful exploitation of the local file read aspect requires the targeted content to be in a valid image or SVG format supported by the library, which adds a layer of complexity but does not eliminate the risk entirely.
The operational impact of this vulnerability extends beyond simple data exposure. In enterprise environments where PDF generation is integrated into document workflows, email attachments, or report generation systems, an attacker could leverage this flaw to perform network reconnaissance against internal services that are not directly exposed to the internet. By observing response codes, timing differences, or error messages returned during the PDF creation process, attackers can infer information about the existence and nature of backend resources. Furthermore, if the server has access to sensitive configuration files or credentials stored in image-like formats, those could potentially be read depending on file permissions and format compatibility. This aligns with CWE-918 which addresses Server-Side Request Forgery (SSRF), where a web application fetches a remote resource without validating the user-supplied URL, leading to potential security breaches involving internal network access or data leakage.
Mitigation strategies primarily involve upgrading to version 2.8.8 of fpdf2, where this issue has been resolved through stricter validation and control over external resource loading. For organizations unable to immediately upgrade, implementing a web application firewall rule set that blocks outbound requests from the PDF generation service to unauthorized domains can provide temporary protection. Additionally, developers should enforce strict input validation on any user-supplied data passed to image-related functions, ensuring that only whitelisted local paths or trusted external URLs are permitted. It is also advisable to run PDF generation services in isolated environments with restricted network access policies to limit the blast radius of potential SSRF exploitation attempts. This approach aligns with MITRE ATT&CK technique T1572 which describes Protocol Tunneling, as attackers may attempt to use legitimate protocols like HTTP or HTTPS to establish covert channels for data exfiltration or command and control communications through the vulnerable application infrastructure.