{"description":"Trending threats, MITRE ATT\u0026CK coverage, and detection metadata. Fed continuously.","feed_url":"https://feed.craftedsignal.io/products/ffmpeg-versions-3.4-through-8.1.2/feed.json","home_page_url":"https://feed.craftedsignal.io/","items":[{"_cs_actors":[],"_cs_cpes":[],"_cs_cves":[{"cvss":7.8,"id":"CVE-2026-65705"}],"_cs_exploited":false,"_cs_has_poc":false,"_cs_poc_references":[],"_cs_products":["FFmpeg versions 3.4 through 8.1.2"],"_cs_severities":["high"],"_cs_tags":["cve","vulnerability","out-of-bounds-write","ffmpeg","video-processing"],"_cs_type":"advisory","_cs_vendors":["FFmpeg Project"],"content_html":"\u003cp\u003eFFmpeg versions 3.4 through 8.1.2 are susceptible to a high-severity out-of-bounds write vulnerability, identified as CVE-2026-65705. This flaw resides within the \u003ccode\u003evf_floodfill\u003c/code\u003e video filter component. Attackers can leverage this by crafting a malicious video stream that features dynamically sized frames. When such a stream is processed by \u003ccode\u003effmpeg\u003c/code\u003e with filtergraph reinitialization explicitly disabled (using the \u003ccode\u003e-reinit_filter 0\u003c/code\u003e command-line argument), the \u003ccode\u003efilter_frame()\u003c/code\u003e function attempts to push flood-fill neighbor data beyond the initially allocated memory boundaries. This heap corruption ultimately leads to a process crash, but depending on the heap layout and system hardening measures, it presents a significant risk for remote code execution. This vulnerability affects a wide range of FFmpeg installations, necessitating immediate patching.\u003c/p\u003e\n\u003ch2 id=\"attack-chain\"\u003eAttack Chain\u003c/h2\u003e\n\u003col\u003e\n\u003cli\u003eAn attacker crafts a malicious video file specifically designed with dynamically sized frames.\u003c/li\u003e\n\u003cli\u003eThe attacker prepares the execution environment, ensuring that the \u003ccode\u003effmpeg\u003c/code\u003e command will process the malicious video with the \u003ccode\u003e-reinit_filter 0\u003c/code\u003e argument, which disables filtergraph reinitialization.\u003c/li\u003e\n\u003cli\u003eThe victim executes the \u003ccode\u003effmpeg\u003c/code\u003e command with the attacker-provided video file and arguments.\u003c/li\u003e\n\u003cli\u003eFFmpeg begins processing the video, and the \u003ccode\u003evf_floodfill\u003c/code\u003e filter allocates a traversal stack based on the dimensions of the initial frames.\u003c/li\u003e\n\u003cli\u003eA subsequent frame, intentionally larger than the initial frames, is processed by the \u003ccode\u003evf_floodfill\u003c/code\u003e filter.\u003c/li\u003e\n\u003cli\u003eThe \u003ccode\u003efilter_frame()\u003c/code\u003e function attempts to write data for flood-fill neighbors beyond the bounds of the original, smaller memory allocation.\u003c/li\u003e\n\u003cli\u003eThis out-of-bounds write corrupts the heap memory of the \u003ccode\u003effmpeg\u003c/code\u003e process.\u003c/li\u003e\n\u003cli\u003eThe heap corruption causes the \u003ccode\u003effmpeg\u003c/code\u003e process to crash, potentially allowing for arbitrary code execution depending on the specific memory layout and system mitigations.\u003c/li\u003e\n\u003c/ol\u003e\n\u003ch2 id=\"impact\"\u003eImpact\u003c/h2\u003e\n\u003cp\u003eSuccessful exploitation of CVE-2026-65705 leads to heap memory corruption within the FFmpeg process, causing it to crash. This results in denial of service for any service or application relying on the vulnerable FFmpeg instance for video processing. Crucially, depending on the specifics of the heap layout and the system's process hardening mechanisms, this vulnerability can be escalated to remote code execution. If exploited, an attacker could gain control over the affected system, leading to data compromise, further network intrusion, or the deployment of additional malware. The broad use of FFmpeg across various applications means the potential impact could span numerous sectors and critical systems.\u003c/p\u003e\n\u003ch2 id=\"recommendation\"\u003eRecommendation\u003c/h2\u003e\n\u003cul\u003e\n\u003cli\u003ePatch CVE-2026-65705 immediately by updating all FFmpeg installations to a version beyond 8.1.2.\u003c/li\u003e\n\u003cli\u003eImplement host-based intrusion detection systems to monitor for unusual \u003ccode\u003effmpeg\u003c/code\u003e process crashes or unexpected terminations.\u003c/li\u003e\n\u003cli\u003eConfigure application whitelisting to prevent the execution of \u003ccode\u003effmpeg\u003c/code\u003e with suspicious command-line arguments like \u003ccode\u003e-reinit_filter 0\u003c/code\u003e from untrusted sources.\u003c/li\u003e\n\u003cli\u003eReview and update system hardening configurations to include exploit mitigation techniques such as Address Space Layout Randomization (ASLR) and Data Execution Prevention (DEP), which can reduce the likelihood of successful code execution.\u003c/li\u003e\n\u003c/ul\u003e\n","date_modified":"2026-07-23T20:21:39Z","date_published":"2026-07-23T20:21:39Z","id":"https://feed.craftedsignal.io/briefs/2026-07-ffmpeg-floodfill-oob-write/","summary":"A critical out-of-bounds write vulnerability exists in FFmpeg versions 3.4 through 8.1.2 within the vf_floodfill video filter, which attackers can exploit by providing a specially crafted, dynamically sized video stream with filtergraph reinitialization disabled via -reinit_filter 0, leading to heap corruption, a process crash, and potentially remote code execution.","title":"Out-of-Bounds Write Vulnerability in FFmpeg vf_floodfill Filter (CVE-2026-65705)","url":"https://feed.craftedsignal.io/briefs/2026-07-ffmpeg-floodfill-oob-write/"}],"language":"en","title":"CraftedSignal Threat Feed - FFmpeg Versions 3.4 Through 8.1.2","version":"https://jsonfeed.org/version/1.1"}