Loading story…
Typora for Windows and Linux contained local file disclosure vulnerabilities that allowed crafted Markdown content to read arbitrary files from a victim system and send them to a remote server. The first flaw, tracked as CVE-2023-2316, affected versions before 1.6.7 and was caused by improper path handling in Typora’s custom Electron protocol handler for typora://app/. An attacker could trigger the issue if a user opened a malicious Markdown file in Typora or pasted attacker-controlled content from a malicious webpage, exposing files such as C:\Windows\win.ini or /etc/passwd.
A second flaw, CVE-2023-2971, was later identified as a patch bypass and incomplete fix for the earlier issue. It affected versions before 1.7.0-dev and abused improper path handling in typora://app/typemark/ to perform path traversal and disclose local files despite Typora’s earlier restriction to paths beginning with typemark. STAR Labs SG researcher Li Jiantao reported both issues, and Typora issued fixes in 1.6.7 and 1.7.0-dev, respectively.
See affected versions and whether adversaries are exploiting it.
5 events from the most recent confirmed update back to the earliest known activity.
STAR Labs published advisories for CVE-2023-2316 and CVE-2023-2971, detailing the original Typora local file disclosure flaw and the later patch bypass. The disclosures included exploitation conditions, affected versions, and proof-of-concept examples showing access to files such as win.ini and /etc/passwd.
Typora fixed CVE-2023-2971 in version 1.7.0-dev after STAR Labs reported the incomplete fix bypass. The patch addressed continued arbitrary local file disclosure risks on Windows and Linux.
STAR Labs researcher Li Jiantao disclosed CVE-2023-2971 to Typora, identifying a path traversal issue in typora://app/typemark/ that bypassed the earlier fix for CVE-2023-2316. The bug could let malicious markdown content read and exfiltrate arbitrary local files if opened or pasted into Typora.
Typora patched CVE-2023-2316, a local file disclosure vulnerability affecting Windows and Linux, in version 1.6.7. The flaw allowed crafted markdown content rendered in Typora to read arbitrary local files via improper path handling in the custom Electron protocol handler.
Typora introduced a mitigation for the original local file disclosure issue in version 1.6.5 by restricting typora://app/ paths to those beginning with typemark. This fix was later found to be incomplete and bypassable.
Vulnerabilities, threat actors, malware, products, organizations, breaches, and observables Mallory has linked to this story. Indicator values are masked here and available in full in the app.
Indicator values are masked on this page. See the values in Mallory Domains, IPs, hashes, and URLs are exportable to your SIEM.
See whether adversaries are exploiting this yet, and where the affected versions run in your environment.
2 references tracked. Mallory keeps watching after this page renders.
Map indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
Multiple newly disclosed vulnerabilities exposed **arbitrary file write** paths through symlink handling and path validation failures in developer and server-side software. In **Crawl4AI** before `0.8.8`, `CVE-2026-56258` affects screenshot and PDF endpoints by combining weak `output_path` validation, symlink following, and a **TOCTOU** race, allowing unauthenticated remote attackers to write files outside the intended directory and potentially reach code execution. In **Cursor Desktop** before `3.0`, `CVE-2026-50549` lets a malicious agent escape the workspace sandbox by forcing path canonicalization to fail and then abusing an in-workspace symlink to write arbitrary files under the user’s privileges, including paths that could enable later unsandboxed execution. A related pattern also appeared in archive extraction and privileged upload handling. The **extract-zip** package is affected by `CVE-2026-56876`, which fails to validate symlink targets inside ZIP archives, enabling path traversal outside the extraction directory and possible arbitrary file read or write when processing untrusted archives. Separately, **pwnlift** disclosed `CVE-2026-56815` in a privileged upload handler that could be abused for arbitrary file write **as root** through direct symlink following and a TOCTOU race; an initial fix was bypassable before a follow-up patch was merged, while one downstream deployment mitigated exposure by removing the privileged `sudo` entry. Together, the disclosures highlight recurring risk from incomplete path canonicalization, unsafe symlink handling, and containment checks in file-write workflows.
Jul 2, 2026
Public proof-of-concept code has been released for three high-severity vulnerabilities in the open-source **File Browser** self-hosted file management application, including **`CVE-2026-73611`**, **`CVE-2026-73612`**, and **`CVE-2026-73613`**. The issues affect File Browser/File Server releases up to **2.63.21** in the broader advisory, while individual fixes were introduced in later versions, including **2.63.22** for the recursive-operations access-control flaw and **2.63.19** as unaffected for the symlink-based deletion issue. `CVE-2026-73612` allows authenticated users to bypass path-based access controls during recursive copy, rename, and delete operations, letting them manipulate restricted descendant files and directories. `CVE-2026-73613` enables arbitrary out-of-scope file deletion through symlink manipulation in the TUS upload cache eviction process, potentially deleting files outside an attacker’s authorized scope with only create permissions. Separately, `CVE-2026-73611` affects JWT-based proxy authentication and can allow expired tokens to be accepted and renewed under certain non-default logout configurations, creating an authentication and security-restrictions bypass risk. Organizations using File Browser have been urged to apply the latest vendor patches immediately.
Aug 14, 2026
Microsoft patched a **Windows Notepad** command-injection vulnerability, **CVE-2026-20841** (CVSS **8.8**), that can lead to **remote code execution** when a user opens a Markdown file in Notepad and clicks a crafted malicious link. The issue is described as improper neutralization of special elements used in a command, enabling an attacker to trigger execution of remote or local payloads in the **security context of the logged-in user**. Public proof-of-concept examples indicate the flaw can be exercised using Markdown `file://` links pointing to executables (e.g., `file://C:/windows/system32/cmd.exe`) and other special URI handlers. The reporting appears as part of a broader weekly “ThreatsDay” roundup that also references other, separate security stories (e.g., AI prompt injection/RCE themes and other malware/exploit items), but the concrete, actionable item consistently detailed is the Notepad Markdown-link RCE and its patch. A separate “Daily Cyber News” post discusses Microsoft releasing fixes for multiple exploited flaws across widely deployed products, but it does not specifically corroborate the Notepad CVE or the Markdown-link exploitation path described in the roundup, making it contextually related to Microsoft patching activity but not the same discrete vulnerability story.
Jun 29, 2026