
Cloud Vulnerability DB
A community-led vulnerabilities database
CVE-2026-77118 is a heap out-of-bounds write vulnerability in the Photo CD (PCD) decoder of GraphicsMagick, classified as High severity. The flaw exists in the DecodeImage() function within coders/pcd.c, where the Huffman delta loop increments an output pointer without bounds checking against the end of heap-allocated luma/chroma plane buffers. All GraphicsMagick versions from 1.0.0 up to (but not including) 1.3.48 are affected, as are corresponding Debian and Fedora packages prior to their respective patched releases. The vulnerability was published on August 20, 2026, with a patch available in GraphicsMagick 1.3.48. It carries a CVSS v4.0 base score of 8.4 (High) (GitHub Advisory).
The root cause is an out-of-bounds write (CWE-787) in DecodeImage() (coders/pcd.c). During Huffman delta decoding of PCD image data, the output pointer q is incremented with q++ after each decoded delta value, but is never validated against the end of the heap-allocated luma or chroma plane buffer. The pointer is only repositioned when a sync marker signals a new plane or row; between sync markers, the number of iterations is controlled entirely by attacker-supplied input. A crafted PCD file can position the pointer near the end of a buffer allocation and then supply a long run of deltas without any intervening sync marker, causing the pointer to walk past the allocation boundary and write arbitrary data to adjacent heap memory. Exploitation requires a user or application to process a malicious PCD file — for example via gm convert, gm identify, or any application linked against libGraphicsMagick (GitHub Advisory).
Successful exploitation allows an attacker to corrupt heap memory beyond the intended buffer boundaries, potentially enabling arbitrary code execution in the context of the process handling the malicious PCD file. The vulnerable system impact includes high confidentiality, integrity, and availability consequences, as heap corruption can be leveraged to overwrite adjacent allocations, control program flow, or cause a crash. Applications that automatically process user-supplied image files (e.g., web services, document converters, or media pipelines using libGraphicsMagick) are at elevated risk, as exploitation requires only that a malicious PCD file be processed (GitHub Advisory).
As of the disclosure date (August 20, 2026), there is no public proof-of-concept exploit and no evidence of in-the-wild exploitation. No threat actor attribution has been reported, and the vulnerability is not listed in the CISA Known Exploited Vulnerabilities (KEV) catalog. Exploitation requires user interaction — specifically, a victim application or user must process a crafted PCD file — which somewhat limits the attack surface compared to fully remote, zero-interaction vulnerabilities. No EPSS score data was available in the provided sources (GitHub Advisory).
q++ loop iterates past the end of the allocated buffer without repositioning.gm convert malicious.pcd output.png or gm identify malicious.pcd, or through any application linked against libGraphicsMagick.gm convert, gm identify, or applications linked against libGraphicsMagick when processing PCD image files.coders/pcd.c or DecodeImage()..pcd files in directories monitored by image processing pipelines, especially files with unusual sizes or structures..pcd files via web upload endpoints, email, or file transfer services to systems running GraphicsMagick-based processing pipelines.Update GraphicsMagick to version 1.3.48 or later, which contains the fix committed at 937cdd9920bd96 in the upstream repository. For Debian, upgrade to 1.4+really1.3.48-1 or later; for Fedora, upgrade to the patched RPM version 1.3.48 or later. As an interim workaround, restrict or disable processing of untrusted PCD files, and disable PCD format support in GraphicsMagick if it is not required by the application. Monitor applications using GraphicsMagick for crashes or unexpected behavior when handling image files (GitHub Advisory).
Source: This report was generated using AI
Free Vulnerability Assessment
Evaluate your cloud security practices across 9 security domains to benchmark your risk level and identify gaps in your defenses.
Get a personalized demo
"Best User Experience I have ever seen, provides full visibility to cloud workloads."
"Wiz provides a single pane of glass to see what is going on in our cloud environments."
"We know that if Wiz identifies something as critical, it actually is."