Explore. Understand.
Stay ahead.
A clearer way to explore known software vulnerabilities. Search CVE records, review severity and CVSS scores, and keep track of the issues that matter to you.
Search the vulnerability database
Find a CVE by ID, keyword, severity or CVSS score.
Latest vulnerability records
matching records · page 961 of 2138
Hugo 0.161.0 placed the Node asset pipelines behind the Node.js permission model so that code running through PostCSS, Babel, or TailwindCSS could not reach the file system outside the project directory. Hugo 0.162.0 added tailwindcss to the AllowChildProcess default in config/security/securityConfig.go, which makes nodePermissionArgs in common/hexec/exec.go append --allow-child-process whenever the tool being launched is named tailwindcss. TailwindCSS loads the site's tailwind.config.js through require at startup,…
CodeWhale versions before 0.8.64 contain an environment variable exposure vulnerability in the js_execution tool that fails to scrub parent process environment variables before spawning Node.js. Attackers can craft malicious JavaScript code executed by the tool to read process.env and leak API keys, cloud credentials, and authentication tokens back to the model context.
CodeWhale versions before 0.8.64 contain a path traversal vulnerability in the image_analyze tool that fails to canonicalize symlinks before reading files. Attackers can create workspace symlinks pointing to external files with image extensions to leak file bytes to the vision endpoint without user approval.
CodeWhale versions >= 0.3.27 and < 0.8.64 (published as deepseek-tui >= 0.3.27 and < 0.8.41 on npm and 0.3.27 through 0.8.41 on crates.io, and as codewhale / codewhale-tui >= 0.8.41 and < 0.8.64) contain an argument injection vulnerability in the git_show tool. The model-supplied rev parameter is passed unvalidated into the git show argv without an --end-of-options sentinel, so a value beginning with --output= is interpreted as a git flag. Because the tool is registered as auto-approved and advertised as read-only,…
CodeWhale versions before 0.8.64 contain an argument injection vulnerability in the git_blame tool that allows attackers to read arbitrary files by injecting git options into the unvalidated rev parameter. Attackers can supply rev values like --contents=/path/to/file to exfiltrate sensitive files such as SSH keys and credentials through the tool output returned to the model.
CodeWhale versions before 0.8.64 fail to properly validate the allow_shell configuration parameter from project config files, allowing attackers to enable arbitrary shell command execution by committing a malicious .codewhale/config.toml file to a repository. When a user clones and opens the repository in CodeWhale, the AI model gains access to exec_shell and task_shell tools, enabling execution of arbitrary shell commands on the victim's machine without explicit user consent.
libmodplug through 0.8.9.1 contains an out-of-bounds read in pat_smplooped in src/load_pat.cpp. The function validates only the upper bound of its sample index against MAXSMP and then subtracts one before indexing the 191-byte static array pat_loops, so an index of zero reads pat_loops[-1], one byte before the array. The index is the smpno field of a parsed MIDI event, which is initialised to zero and only later overwritten from a program-change parameter, so an event reaching the note test before an instrument is …
CodeWhale versions before 0.8.64 fail to validate file paths in the project config instructions field, allowing attackers to read arbitrary files on the victim's system. A malicious .codewhale/config.toml file in a cloned repository can specify paths outside the workspace that are read and injected into the AI system prompt for exfiltration.
CodeWhale (packages codewhale / codewhale-tui) versions >= 0.8.41 and < 0.8.64 contain a remote code execution vulnerability in the rlm_eval tool. The tool's approval_requirement() returns ApprovalRequirement::Auto, which the engine treats as 'never prompt,' causing arbitrary model-supplied Python code to run in a python3 interpreter without consulting the user's configured --approval-policy and without any approval prompt or audit step. An attacker can induce the agent to execute arbitrary code via prompt injectio…
CodeWhale versions >= 0.3.10 and < 0.8.64 (published as deepseek-tui >= 0.3.10 and < 0.8.41 on npm and 0.3.10 through 0.8.41 on crates.io, and as codewhale / codewhale-tui >= 0.8.41 and < 0.8.64) contain a vulnerability in the exec_shell_interact (alias exec_interact) tool, whose approval_requirement returns ApprovalRequirement::Auto. This overrides the default Required approval for code-executing tools, so LLM-controlled stdin is written into an already-approved long-running interactive shell (e.g., a python3 -i R…
CodeWhale before 0.8.64 contains a server-side request forgery bypass vulnerability in DNS pinning logic that fails to prevent time-of-check-time-of-use attacks. Attackers can manipulate DNS responses to fail initial resolution checks and succeed on secondary requests, allowing requests to internal IP addresses and bypassing SSRF mitigations.
Joomla! Core - [20260810] - Unrestricted uploads of SHTML files in Joomla 1.0.0-5.4.7, 6.0.0-6.1.2 - The default list of dangerous files did not include SHTML files. On servers that executed these files, that could lead to code execution.
Joomla! Core - [20260808] - Improper ACL checks for batch copy actions in Joomla 4.0.0-5.4.7, 6.0.0-6.1.2 - An improper access check allows unauthorized users to perform copy batch operations on uneditable items.
Joomla! Core - [20260807] - MFA Authentication Bypass in Joomla 4.0.0-5.4.7 and 6.0.0-6.1.2 - Insufficient state checks lead to a vector that allows to bypass 2FA checks.
Joomla! Core - [20260806] - Improper ACL checks for category webservice endpoints in Joomla 4.0.0-5.4.7, 6.0.0-6.1.2 - An improper access check allows unauthorized users to create categories via webservices endpoints.
Joomla! Core - [20260803] - Inconsistent ACL checks for mutating webservice endpoints in Joomla 4.0.0-5.4.7, 6.0.0-6.1.2 - An improper access check allows unauthorized users to perform mutation actions in webservice endpoints, where the same mutation was restricted in the backend UI.
mise manages dev tools like node, python, cmake, and terraform. Prior to 2026.7.1, release tar archives record mise/bin/mise with user and group ID 1001 and packaging/standalone/install.envsubst extracts and moves it without normalizing ownership, allowing a local user with those IDs to replace a root-installed executable, especially when MISE_INSTALL_PATH targets a shared location such as /usr/local/bin. This issue is fixed in version 2026.7.1.
A server-side request forgery (SSRF) vulnerability was found in AWX's webhook status callback mechanism. When processing GitHub pull request webhooks, AWX extracts the status callback URL (pull_request.statuses_url) from the incoming webhook payload without validating the target host against the expected Git provider. This URL is persisted in job extra variables and later used to send authenticated status updates. A user with admin role on a webhook-enabled job template can read the template's webhook signing key, …
Trivy is a security scanner. Prior to 0.72.0, plugin manifest metadata is used by pkg/plugin/manager.go to construct paths under ~/.trivy/plugins without confining plugin names to that root, allowing an attacker who persuades a user to install or run a malicious plugin to write the manifest and plugin binary to arbitrary user-writable paths, while plugins from the official Trivy plugin index are not affected. This issue is fixed in version 0.72.0.
Dragonfly is an in-memory data store built for modern application workloads. Prior to 1.40.0, CMS.INITBYDIM and CMS.INITBYPROB accept dimensions whose width times depth times sizeof(int64_t) overflows in src/core/cms.cc, allocating an undersized counter buffer while CMS.INCRBY and CMS.QUERY use the unbounded dimensions, which allows an unauthenticated remote client to corrupt or disclose adjacent heap memory and crash the server. This issue is fixed in version 1.40.0.
Kestra is an open-source, event-driven orchestration platform. Prior to 1.3.24, Kestra's custom Markdown parser in ui/src/utils/markdown_plugins/link.ts allows a user with permission to create or update a Flow description to inject JavaScript event-handler attributes through the custom [[link]] syntax, causing stored cross-site scripting when another user opens the description or information panel in the Flow list. This issue is fixed in version 1.3.24.
Vvveb is a powerful and easy to use CMS with page builder to build websites, blogs or ecommerce stores. Prior to 1.0.8.4, Vvveb backend comment operations allow a low-privileged Author to manage comments under another Author's posts. The admin/controller/content/comment.php and admin/controller/content/comments.php controllers and the admin/sql/sqlite/comment.sql queries accept a caller-controlled comment_id without verifying comment.post_id against post.admin_id for the current admin_id. An attacker can read pendi…
Vvveb is a powerful and easy to use CMS with page builder to build websites, blogs or ecommerce stores. Prior to 1.0.8.4, Vvveb backend post operations allow a low-privileged Author to access posts owned by another Author. The admin/controller/content/posts.php controller permits filter[admin_id] to replace the server-selected admin_id restriction and accepts a caller-controlled post_id for duplicate and delete actions, while admin/sql/sqlite/post.sql does not consistently enforce post.admin_id. An attacker can vie…
Vvveb is a powerful and easy to use CMS with page builder to build websites, blogs or ecommerce stores. Prior to 1.0.8.4, Vvveb backend digital asset operations allow a low-privileged Vendor to access digital assets linked to another Vendor's products. The admin/controller/product/digital-asset.php and admin/controller/product/digital-assets.php controllers and the admin/sql/sqlite/digital_asset.sql data queries use a caller-controlled digital_asset_id without consistently enforcing the current admin_id ownership b…
The HTTPPasswordMgr class in the urllib.request module, along with its subclasses HTTPPasswordMgrWithDefaultRealm and HTTPPasswordMgrWithPriorAuth, did not take the URL scheme into account when matching stored credentials against a requested URL. Credentials added for an https:// URL were also used for requests to the same host over http://, so an attacker able to redirect or downgrade a client to plain HTTP (for example, via an HTTPS-to-HTTP redirect or an on-path position) could capture credentials in cleartext. …