Overview 267 matches · 13 in Waratek scope · 6 protected · 3 exploited · 2 CISA KEV · 5 public PoC · 2 EPSS ≥ 0.5
267matches
13in Waratek scope4.9%
6protected2.2%
3exploited1.1%
2CISA KEV0.7%
5public PoC1.9%
2EPSS ≥ 0.50.7%
Protected by Waratek
6
2.2%
Rule in development
0
0.0%
Mitigation candidate
0
0.0%
No exploit published
7
2.6%
Outside Waratek scope
5
1.9%
Not applicable
249
93.3%
Withdrawn CVE
0
0.0%
blocked by ARMR today
6
2.2%
partly blocked
0
0.0%
not blocked
261
97.8%
not established
0
0.0%
unrecorded
0
0.0%
split by
peak 205 / year
20174
Not applicable4
Critical2
High1
Medium1
not blocked4
No public exploit4
20188
Protected by Waratek4
No exploit published1
Not applicable3
Critical3
High4
Medium1
blocked by ARMR today4
not blocked4
Working exploit published1
Proof of concept only1
No public exploit6
20191
Protected by Waratek1
Critical1
blocked by ARMR today1
No public exploit1
20202
Outside Waratek scope1
Not applicable1
Critical1
High1
not blocked2
No public exploit2
20218
No exploit published1
Outside Waratek scope4
Not applicable3
Critical1
High3
Medium4
not blocked8
No public exploit8
20225
Protected by Waratek1
No exploit published1
Not applicable3
High2
Medium3
blocked by ARMR today1
not blocked4
Proof of concept only1
Forecast only1
No public exploit3
20239
No exploit published1
Not applicable8
Critical2
High5
Medium2
not blocked9
Proof of concept only4
No public exploit5
202411
Not applicable11
Critical1
High3
Medium6
Low1
not blocked11
Exploited2
Proof of concept only2
No public exploit7
202514
Not applicable14
Critical2
High5
Medium5
Low2
not blocked14
Proof of concept only6
No public exploit8
2026205
No exploit published3
Not applicable202
Critical33
High95
Medium60
Low17
not blocked205
Exploited1
Working exploit published2
Proof of concept only72
No public exploit130
2017201820192020202120222023202420252026
| CVE | Severity | Published▾ | Status | Exploitation | Description |
|---|---|---|---|---|---|
| CVE-2026-72779 | High | 2026-08-11 | Not applicable | No public exploit | Craft CMS 5.0.0-RC1 before 5.10.6 and 4.0.0-RC1 before 4.18.2 contain an arbitrary file read vulnerability. The create() Twig function restricts class instantiation using a 5-entry blocklist that does not include SplFileObject, allowing an authenticated administrator (with allowAdminChanges=true) to configure a malicious entry type title or URI format that instantiates SplFileObject in a non-sandboxed template context. When a user subsequently creates an entry in the affected section, arbitrary files on the server (such as .env containing the security key and database credentials) are read and rendered as entry titles. |
| CVE-2026-71259 | High | 2026-08-05 | Not applicable | No public exploit | ESPHome through 2026.7.0-dev contains an operator-precedence bug in the cv.url validator in esphome/config_validation.py. Because binds tighter than , any file: URI passes validation regardless of netloc. This validator gates the field of the external_components YAML directive's git source schema, which is passed to (git supports file:// natively). |
| CVE-2026-17630 | High | 2026-08-05 | Not applicable | No public exploit | IBM Langflow OSS 1.0.0 through 1.10.3 could allow a remote attacker to execute arbitrary code due to improper validation of configuration parameters. |
| CVE-2026-70470 | Critical | 2026-08-04 | Not applicable | Proof of concept only | Flowise is a drag & drop user interface to build a customized large language model flow. Prior to 3.1.3, Flowise validatePythonCodeForDataFrame in packages/components/src/pythonCodeValidator.ts can be bypassed with Unicode homoglyph identifiers, allowing arbitrary Python execution inside Pyodide and full OS command execution on the Flowise host via Pyodide js module interop. The validator gates pyodide.runPythonAsync in packages/components/nodes/agents/CSVAgent/CSVAgent.ts and packages/components/nodes/agents/AirtableAgent/AirtableAgent.ts with an ASCII word-boundary blacklist. JavaScript regex word boundaries are ASCII-only, while Python 3 NFKC-normalizes identifiers at parse time, so homoglyph forms such as __cl𝐚ss__, __subcl𝐚sses__, __b𝐚se__, and __b𝐮iltins__ bypass the blacklist and are parsed as their ASCII equivalents. This issue is fixed in version 3.1.3. |
| CVE-2026-69263 | High | 2026-08-04 | Not applicable | Proof of concept only | Flowise is a drag & drop user interface to build a customized large language model flow. Prior to 3.1.3, the mitigation for CVE-2025-8943 blocked -y and --yes flags on npx, but packages/components/nodes/tools/MCP/core.ts denied only PATH, LD_LIBRARY_PATH, DYLD_LIBRARY_PATH, and NODE_OPTIONS by exact environment-variable name. Because npm reads configuration from npm_config_* variables, setting npm_config_yes=true reproduced --yes behavior without using a blocked flag, causing npx to auto-install and execute the named package when a Custom MCP server launched. This issue is fixed in version 3.1.3. |
| CVE-2026-18174 | Medium | 2026-07-29 | Not applicable | No public exploit | @fastify/forwarded resolves client addresses from the X-Forwarded-For header. In versions before 3.0.2, when the header contains two or more comma separated entries, the parser trims only space characters and does not strip horizontal tabs, even though RFC 7230 defines optional whitespace as both space and tab. As a result, an entry padded with a tab keeps the literal tab in the resolved address string. Applications that make exact string match security decisions on the resolved client IP, such as an allowlist, a blocklist, a per IP rate limit key, or audit log correlation, can be evaded because the tab corrupted string no longer matches the expected value. This does not cross the trust boundary, since a tab corrupted string is not a valid IP and cannot be mistaken for a trusted proxy. The issue is fixed in @fastify/forwarded 3.0.2. |
| CVE-2026-50251 | Medium | 2026-07-22 | Not applicable | No public exploit | In NLnet Labs Unbound up to and including version 1.25.1, when 'unwanted-reply-threshold' is enabled (set to any value greater than zero), glue records of 0.0.0.0/::0 can short-circuit Unbound, on systems that can direct such traffic, by issuing DNS queries and receiving seemingly unwanted replies since the remote IP does not match the original source IP of 0.0.0.0/::0. This behavior keeps on looping for the glue records and pushing the counter to the configured 'unwanted-reply-threshold' that triggers a defensive cache clear. A malicious actor who controls a delegation that returns in-bailiwick glue of 0.0.0.0/::0 can drive the counter to the limit of 'unwanted-reply-threshold' to the threshold and trigger a cache clean of the message and rrset caches; at will, indefinitely, without sending a single spoofed packet. The iterator uses the 0.0.0.0/::0 glue, and a system that can route this (e.g., Linux kernel routes the datagram over loopback), Unbound's own listener answers from 127.0.0.1. Because of the mismatch of 0.0.0.0 and 127.0.0.1, in this example, Unbound accounts the reply as an unwanted (probably spoofed) answer. The counter resets to zero on every cache flush, so the attack loops forever. |
| CVE-2026-47392 | Critical | 2026-07-21 | Not applicable | Proof of concept only | PraisonAI is a multi-agent teams system. Prior to version 4.6.40 of PraisonAI, corresponding to version 1.6.40 of praisonaiagents, `execute_code()` in `praisonaiagents/tools/python_tools.py` (v1.6.37, subprocess sandbox mode) can be fully bypassed using `print.__self__` to retrieve the real Python `builtins` module, from which `__import__` can be extracted via `vars()` and runtime string construction. This achieves arbitrary OS command execution on the host, completely defeating the sandbox. This is a novel bypass that survives all patches for CVE-2026-39888 (frame traversal), CVE-2026-34938 (str subclass), and CVE-2026-40158 (`type.__getattribute__` trampoline). PraisonAI version 4.6.40 and praisonaiagents version 1.6.40 contain an updated fix. |
| CVE-2026-63108 | High | 2026-07-20 | Not applicable | Proof of concept only | Roo Code through 3.54.0 contains a command injection vulnerability in the auto-approve execute feature that allows attackers to bypass allowlist/denylist enforcement by nesting command substitutions inside parameter expansion defaults. The command parser in parse-command.ts replaces parameter expansions with opaque placeholders before extracting command substitutions, causing the containsDangerousSubstitution guard to miss nested payloads, which are then auto-approved based on the outer allowlisted command prefix and executed by the shell via execa, enabling arbitrary command execution. |
| CVE-2026-16129 | Low | 2026-07-18 | Not applicable | Proof of concept only | A vulnerability has been found in princezuda SafestClaw up to 4.2.4. This vulnerability affects the function ShellAction._validate_command of the file src/safestclaw/actions/shell.py of the component Built-in Web Interface. Such manipulation leads to incomplete blacklist. An attack has to be approached locally. The exploit has been disclosed to the public and may be used. The presence of this vulnerability remains uncertain at this time. The project maintainer explains: "On paper you're correct, this is a vulnerability. In practice, nothing your AI generated shows how it makes users vulnerable. It's open source. Someone can mod the shell allow list or remove that system. Present an actual poc that shows a threat to users." |
| CVE-2026-62203 | High | 2026-07-17 | Not applicable | No public exploit | OpenClaw versions before 2026.6.6 contain an environment variable filtering vulnerability in host exec that fails to properly sanitize rustup startup variables. Attackers with lower-trust caller access or configured input paths can execute or persist actions beyond their intended authorization level. |
| CVE-2026-13448 | Critical | 2026-07-17 | Not applicable | No public exploit | IBM Langflow OSS 1.0.0 through 1.10.1 Lanflow OSS contains an unauthenticated remote code execution vulnerability in the public flow build endpoint ( /api/v1/build_public_tmp/{flow_id}/flow ). The vulnerability stems from an incomplete denylist in the validate_public_flow_no_code_execution() function that fails to block several code-execution agent components including OpenDsStarAgent, CodeActAgentSmolagents, and CSVAgent. |
| CVE-2026-52888 | Medium | 2026-07-15 | Not applicable | Proof of concept only | NocoBase is an AI-powered no-code/low-code platform for building business applications and enterprise solutions. In 2.0.59 and earlier, NocoBase @nocobase/plugin-collection-sql used the checkSQL() function in packages/plugins/@nocobase/plugin-collection-sql/src/server/utils.ts with an incomplete keyword blacklist that did not restrict PostgreSQL system catalog tables such as pg_shadow, pg_roles, and pg_stat_activity, allowing an admin-role user to read password hashes and database metadata through the SQL Collection feature. This vulnerability is fixed in 2.1.0-alpha.46. |
| CVE-2026-48736 | Medium | 2026-07-14 | Not applicable | No public exploit | Symfony is a PHP framework for web and console applications and a set of reusable PHP components. From 5.4.0 to 5.4.53, 6.4.41, 7.4.13, and 8.0.13, NoPrivateNetworkHttpClient and IpUtils::PRIVATE_SUBNETS omitted IPv6 transition prefixes such as 6to4, NAT64, Teredo, and IPv4-compatible IPv6, allowing attacker-supplied URLs to represent private IPv4 targets in forms that IpUtils::isPrivateIp() did not block. This issue is fixed in versions 5.4.53, 6.4.41, 7.4.13, and 8.0.13. |
| CVE-2026-45753 | Low | 2026-07-14 | Not applicable | No public exploit | Symfony is a PHP framework for web and console applications and a set of reusable PHP components. From 6.1.0-BETA1 until 6.4.40, 7.4.12, and 8.0.12, UrlAttributeSanitizer::getSupportedAttributes() omits URL-valued attributes including action, formaction, poster, and cite, so configurations that admit those attributes can leave javascript: URIs unsanitized and enable XSS when the resulting HTML is rendered or a victim submits a form or clicks a button. This issue is fixed in versions 6.4.40, 7.4.12, and 8.0.12. |
| CVE-2026-45066 | Low | 2026-07-14 | Not applicable | No public exploit | Symfony is a PHP framework for web and console applications and a set of reusable PHP components. From 6.1.0-BETA1 until 6.4.40, 7.4.12, and 8.0.12, HtmlSanitizer URL sanitization can allow off-allowlist URLs through allowLinkHosts() or allowMediaHosts() because UrlSanitizer::parse() follows RFC 3986 while browsers follow WHATWG URL parsing, and because <area href> is checked against the media policy rather than the link policy. This issue is fixed in versions 6.4.40, 7.4.12, and 8.0.12. |
| CVE-2026-15625 | Low | 2026-07-14 | Not applicable | Proof of concept only | A vulnerability was found in nextlevelbuilder GoClaw 3.11.3. Affected by this issue is the function ExecApprovalManager.CheckCommand of the file internal/tools/exec_approval.go. The manipulation results in incomplete blacklist. The attack can be executed remotely. The exploit has been made public and could be used. |
| CVE-2026-62200 | High | 2026-07-13 | Not applicable | No public exploit | OpenClaw versions before 2026.6.6 contain a flaw in host exec environment filtering that could allow Git ext transport to be abused. When the affected feature is enabled and reachable, a lower-trust caller or configured input path could execute or persist actions beyond the caller's intended authorization. |
| CVE-2026-62199 | High | 2026-07-13 | Not applicable | No public exploit | OpenClaw versions before 2026.6.6 contain a flaw in host exec environment filtering that can miss interpreter startup variables. When the affected feature is enabled and reachable, a lower-trust caller or configured input path can supply crafted environment variables to execute or persist actions beyond the caller's intended authorization. |
| CVE-2026-59929 | Medium | 2026-07-08 | Not applicable | No public exploit | Mistune is a Python Markdown parser with renderers and plugins. Prior to 3.3.0, the safe_url filter in src/mistune/renderers/html.py blocks only javascript:, vbscript:, file:, and data: schemes, allowing legacy or chained schemes such as feed:, view-source:, jar:, livescript:, mocha:, ms-its:, mk:, and res: to reach rendered href and src attributes and potentially execute script in affected user agents. This issue is fixed in version 3.3.0. |
| CVE-2026-59261 | High | 2026-07-08 | Not applicable | No public exploit | OpenClaw before 2026.5.28 contains a credential exposure vulnerability where workspace dotenv files can override provider credentials. Attackers with lower-trust access to configured input paths can expose sensitive data and credentials that should remain within trusted boundaries. |
| CVE-2026-55830 | High | 2026-07-08 | Not applicable | Proof of concept only | RestrictedPython is a tool that helps to define a subset of the Python language which allows to provide a program input into a trusted environment. Prior to 8.3, check_function_argument_names() rejected protected guard hook names for regular, variadic, and keyword-only arguments but omitted positional-only arguments, allowing __getattr__, _getitem_, _write_, or _print_ to be shadowed by a local parameter and bypass the embedding application's access policy. This issue is fixed in version 8.3. |
| CVE-2026-14534 | High | 2026-07-04 | Not applicable | Proof of concept only | Trail of Bits fickling versions up to and including 0.1.10 do not include the Python standard library modules _posixsubprocess, site, and atexit in the UNSAFE_IMPORTS denylist (fickle.py). Because these modules are absent from the denylist, fickling's check_safety() function returns LIKELY_SAFE with zero findings for pickle payloads that invoke dangerous functions including _posixsubprocess.fork_exec (C-level process spawner capable of executing arbitrary binaries), site.execsitecustomize (executes arbitrary site customization code), and atexit._run_exitfuncs (triggers all registered exit handler callbacks). The fickling.load() API chains check_safety() into pickle.loads() as an explicit security gate; a LIKELY_SAFE verdict causes the payload to be deserialized and executed. This shares the same root cause as CVE-2026-22607 (cProfile), CVE-2025-67748 (pty), and CVE-2025-67747 (marshal/types). OvertlyBadEvals does not flag these modules because they are standard library imports. UnsafeImports does not flag them because they are not in the denylist. The UnusedVariables heuristic is defeated by the SETITEMS opcode pattern. |
| CVE-2026-56777 | Medium | 2026-06-30 | Not applicable | No public exploit | n8n before 2.25.7 and 2.26.x before 2.26.2 contains an abstract syntax tree (AST) security validator bypass in the Python Code node. An authenticated user with permission to create or modify workflows containing a Python Code node can bypass the validator and access the task executor module namespace. The issue only affects self-hosted instances where the Python Task Runner is enabled; where N8N_BLOCK_RUNNER_ENV_ACCESS is configured to allow it, this can disclose environment variables accessible to the task runner process. |
| CVE-2025-71355 | High | 2026-06-30 | Not applicable | Proof of concept only | Picklescan before 0.0.25 fails to detect unsafe global functions in the Numpy library, allowing attackers to bypass static analysis and execute arbitrary code during deserialization. Attackers can craft malicious pickle files using numpy.testing._private.utils.runstring within the reduce method to import dangerous libraries like os and execute arbitrary OS commands when the pickle file is loaded. |
| CVE-2026-49869 | Critical | 2026-06-26 | Not applicable | Exploited | Kestra is an open-source, event-driven orchestration platform. Prior to 1.0.45 and 1.3.21, AuthenticationFilter in Kestra OSS uses request.getPath().endsWith("/configs") to whitelist the public configuration endpoint from Basic Auth. Because the check is a suffix match rather than an exact path match, any API path whose last segment is configs bypasses authentication entirely. An unauthenticated remote attacker can exploit this to create and execute arbitrary workflows without credentials. Because Kestra ships with script execution plugins (plugin-script-shell, plugin-script-python, etc.) enabled by default, this directly results in unauthenticated Remote Code Execution as root inside the Kestra worker container. This vulnerability is fixed in 1.0.45 and 1.3.21. |
| CVE-2026-57234 | Low | 2026-06-25 | Not applicable | No public exploit | Nokogiri is an open source XML and HTML library for the Ruby programming language. Prior to 1.19.4, the NONET parse option, which Nokogiri turns on by default for Nokogiri::XML::Schema (see CVE-2020-26247), was not correctly enforced on the JRuby implementation. As a result, a schema parsed with default options could still cause external resources to be fetched over the network, potentially enabling SSRF or XXE attacks. This vulnerability is fixed in 1.19.4. |
| CVE-2026-54090 | High | 2026-06-25 | Not applicable | Proof of concept only | File Browser is a file managing interface for uploading, deleting, previewing, renaming, and editing files within a specified directory. Prior to 2.33.8, when a shell interpreter is configured (e.g. /bin/sh -c), the command allowlist can be bypassed through shell metacharacters. The allowlist validates only the first token of user input, but the entire raw string is handed to the shell — semicolons, pipes, backticks, and $() all work to chain arbitrary commands after a permitted one. This vulnerability is fixed in 2.33.8. |
| CVE-2026-54070 | High | 2026-06-24 | Not applicable | No public exploit | SiYuan is an open-source personal knowledge management system. Prior to 3.7.0, renderPackageREADME in kernel/bazaar/readme.go renders a Bazaar package README from Markdown to HTML with the lute engine and SetSanitize(true). The lute sanitizer is an event-handler blocklist: allowAttr rejects only attribute names present in a fixed eventAttrs map copied from the w3schools legacy handler list. That map omits modern event handlers. onpointerover, onpointerdown, onauxclick, onbeforetoggle, onfocusin, onanimationstart, and ontransitionend are not in the list, so the sanitizer passes them through verbatim on any tag. The frontend assigns the rendered HTML to mdElement.innerHTML in app/src/config/bazaar.ts with no client-side DOMPurify on this path, into a normal element in the main document (no iframe, no sandbox). The kernel sends no Content-Security-Policy, X-Frame-Options, or X-Content-Type-Options header on any response, so an inline handler runs when its event fires. The README is rendered when an Administrator opens a package in Settings → Marketplace, after the one-time marketplace trust consent. Install is not required. Result: a third-party Bazaar package author runs JavaScript in the Administrator's authenticated SiYuan origin when the Administrator views and interacts with the package listing, and gains full control of the workspace. This vulnerability is fixed in 3.7.0. |
| CVE-2026-53944 | Medium | 2026-06-24 | Not applicable | No public exploit | Ghost is a Node.js content management system. From 6.0.9 until 6.21.1, when making an external request, it is possible to bypass the IP filter that ensures the request isn't going to an internal service using an IPv6 literal which maps to a private IPv4 address. This vulnerability is fixed in 6.21.1. |
| CVE-2026-47389 | High | 2026-06-24 | Not applicable | No public exploit | Mastodon is a free, open-source social network server based on ActivityPub. Prior to 4.5.10, 4.4.17, and 4.3.23, when using Ruby versions older than 3.4, PrivateAddressCheck.private_address? returns false for IPv4-mapped IPv6 addresses (::ffff:a.b.c.d) corresponding to some private IPv4 addresses, depending on Ruby version, this can include loopback, RFC1918 private networks, and link-local space. An attacker who controls DNS for any domain can publish an AAAA record with such a mapped address; any outbound HTTP fetch Mastodon performs against that hostname then opens a real TCP connection to the underlying IPv4 address, including 127.0.0.1 and cloud-metadata endpoints such as 169.254.169.254. This vulnerability is fixed in 4.5.10, 4.4.17, and 4.3.23. |
| CVE-2026-56315 | Critical | 2026-06-23 | Not applicable | Proof of concept only | picklescan before 1.0.4 fails to block at least seven Python standard library modules (including uuid, _osx_support, _aix_support, _pyrepl.pager, and imaplib) exposing eight functions that provide direct arbitrary command execution. Attackers can craft malicious pickle files importing these unblocked modules to achieve remote code execution while bypassing picklescan's safety validation entirely. |
| CVE-2026-54513 | High | 2026-06-23 | No exploit published | No public exploit | jackson-databind contains the general-purpose data-binding functionality and tree-model for Jackson Data Processor. From 2.10.0 until 2.18.8, 2.21.4, and 3.1.4, BasicPolymorphicTypeValidator.Builder.allowIfSubTypeIsArray() allowlists any array type based only on clazz.isArray(), without validating the array's component (element) type against the configured allowlist. A PTV built with allowIfSubTypeIsArray() plus an explicit concrete-type allowlist therefore still permits EvilType[] even though EvilType is not allowlisted. When Jackson deserializes the elements and no per-element type IDs are present, it instantiates the component type directly with no further PTV check, bypassing the allowlist. This vulnerability is fixed in 2.18.8, 2.21.4, and 3.1.4. |
| CVE-2026-54512 | High | 2026-06-23 | No exploit published | Proof of concept only | jackson-databind contains the general-purpose data-binding functionality and tree-model for Jackson Data Processor. From 2.10.0 until 2.18.8, 2.21.4, and 3.1.4, jackson-databind's PolymorphicTypeValidator (PTV) is the primary safety mechanism guarding polymorphic deserialization. When polymorphic typing is enabled and a type identifier contains generic parameters (i.e. the type ID string contains <), DatabindContext._resolveAndValidateGeneric() validates only the raw container class name (the substring before <) against the configured PTV. If the container type is approved, the method parses the full canonical type string via TypeFactory.constructFromCanonical() and returns the fully parameterized type without ever validating the nested type arguments against the PTV. The nested type arguments are then resolved, instantiated, and populated as beans during deserialization. An attacker who controls the type ID can therefore place a denied class as a generic type parameter of an allowed container — for example java.util.ArrayList<com.evil.Gadget> when only java.util.ArrayList is allow-listed. The container passes the PTV check; com.evil.Gadget is loaded via Class.forName(name, true, loader), instantiated, and its properties are set from attacker-controlled JSON. This completely bypasses an explicitly configured PTV allow-list. This vulnerability is fixed in 2.18.8, 2.21.4, and 3.1.4. |
| CVE-2025-71351 | High | 2026-06-21 | Not applicable | Proof of concept only | picklescan before 0.0.25 fails to detect malicious pickle files that use timeit.timeit() in the __reduce__ method, allowing remote code execution. Attackers can craft pickle files that import dangerous libraries like os and execute arbitrary system commands, which evade picklescan detection and execute when pickle.load() is called. |
| CVE-2026-55743 | Critical | 2026-06-17 | Not applicable | No public exploit | The shell tool command allowlist in the SecurityPolicy of OpenHuman desktop agent through 0.54.0 (default Supervised security policy) can be bypassed to execute arbitrary OS commands with the privileges of the desktop user. |
| CVE-2026-53873 | Critical | 2026-06-17 | Not applicable | Proof of concept only | picklescan before 1.0.4 contains an incomplete blocklist for the profile module that fails to block the module-level profile.run() function, allowing attackers to achieve arbitrary code execution via exec(). Attackers can craft malicious pickle files calling profile.run(statement) to execute arbitrary Python code while picklescan reports zero security issues. |
| CVE-2026-44587 | Medium | 2026-06-17 | Not applicable | Proof of concept only | CarrierWave is a framework to upload files from Ruby applications. In versions prior to 2.2.7 and 3.1.3, the content_type_denylist check fails to escape regex metacharacters in string entries, causing the denylist to silently not match the content types it is intended to block. In lib/carrierwave/uploader/content_type_denylist.rb:57, denylist entries are interpolated directly into a regex without Regexp.quote or anchoring, so an entry such as image/svg+xml becomes the pattern /image\/svg+xml/, in which + is treated as a quantifier rather than a literal character and therefore never matches the real MIME type image/svg+xml. This is inconsistent with the allowlist implementation, which correctly applies both Regexp.quote and a \A anchor. Other content types containing regex metacharacters, such as application/xhtml+xml, are affected as well. As a result, any application that relies on content_type_denylist to block image/svg+xml, most commonly to prevent stored XSS, is silently unprotected. An attacker can upload an SVG file containing arbitrary JavaScript; if the application serves that SVG inline from its own origin, the script executes in the victim's browser, resulting in stored XSS. This issue has been fixed in versions 2.2.7 and 3.1.3. |
| CVE-2025-71323 | Critical | 2026-06-17 | Not applicable | Proof of concept only | picklescan before 0.0.33 fails to block the ctypes module, allowing attackers to achieve remote code execution by invoking direct syscalls and accessing raw memory. Attackers can craft malicious pickle files using ctypes.WinDLL to load kernel32.dll and execute arbitrary commands, bypassing sandbox protections and gadget chain detection. |
| CVE-2025-71320 | Critical | 2026-06-17 | Not applicable | Proof of concept only | picklescan before 0.0.33 contains an incomplete deny-list that fails to block pydoc.locate and operator.methodcaller functions, allowing attackers to bypass security checks. Remote attackers can craft malicious pickle files using these unblocked functions to achieve arbitrary code execution when the pickle is deserialized. |
| CVE-2026-53864 | High | 2026-06-16 | Not applicable | No public exploit | OpenClaw before 2026.5.26 contains an insufficient sanitization vulnerability in the host environment sanitizer that allows Node.js control variables to bypass validation. Attackers with access to workspace .env files, tool environment overrides, or skill environment blocks can pass malicious Node.js control variables to influence child processes or coverage output paths. |
| CVE-2026-53861 | Medium | 2026-06-16 | Not applicable | No public exploit | OpenClaw before 2026.5.6 contains an allowlist bypass vulnerability in the macOS Swift exec feature that misses combined POSIX inline-command flags. Attackers can execute shell content outside the intended allowlist check by using combined flag forms, potentially allowing unauthorized command execution depending on operator configuration. |
| CVE-2026-53855 | High | 2026-06-16 | Not applicable | No public exploit | OpenClaw before 2026.4.2 contains an inline-eval bypass vulnerability allowing authenticated operators to weaken strict allowlist checks via shell positional parameters. Attackers can combine allowlisted tools with shell positional arguments to place inline-eval content in shell carriers outside intended allowlist rules, enabling execution of unapproved shell-provided content. |
| CVE-2026-53848 | Low | 2026-06-16 | Not applicable | No public exploit | OpenClaw before 2026.5.26 contains an exec allowlist bypass vulnerability allowing authenticated operators to execute wrapper-level side effects outside allowlisted command intent. Attackers can craft command requests that bypass allowlist validation by leveraging transparent command wrappers to perform unintended operations. |
| CVE-2026-53836 | High | 2026-06-12 | Not applicable | No public exploit | OpenClaw before 2026.5.12 contains an allowlist bypass vulnerability in PowerShell encoded-command handling that allows attackers to execute encoded commands using abbreviated flag aliases not recognized by the allowlist parser. Remote authenticated operators can bypass execution allowlist checks by using unrecognized encoded-command alias forms to execute arbitrary PowerShell content. |
| CVE-2026-48557 | High | 2026-05-29 | Not applicable | No public exploit | Spatie Laravel Media Library before version 11.23.0 contains a file upload restriction bypass in FileAdder::defaultSanitizer(). The sanitizer checks only the final filename suffix, allowing double-extension filenames such as shell.php.jpg to bypass the blocklist, with pathinfo() preserving inner .php stems in saved filenames. The blocklist also omits executable extensions including .php6, .shtml, and .htaccess. The double-extension bypass requires a legacy Apache AddHandler configuration to achieve PHP execution; the incomplete blocklist bypass does not. |
| CVE-2026-44287 | Medium | 2026-05-29 | Not applicable | Proof of concept only | FastGPT is an AI Agent building platform. Prior to 4.15.0-beta1, the JavaScript sandbox worker at projects/code-sandbox/src/pool/worker.ts:356 blocks dynamic import() with the regex /\bimport\s*\(/.test(code). JavaScript syntax accepts a block comment between import and (; the regex matches only ASCII whitespace, and the bytes /, *, *, / are not in the \s character class. The payload import/**/("child_process") parses as a syntactically valid dynamic import that the regex does not detect. Because import() is not wrapped by the safeRequire Proxy (which only proxies require), the attacker loads child_process and calls execSync - arbitrary command execution as uid=100(sandbox) inside the sandbox container. This vulnerability is fixed in 4.15.0-beta1. |
| CVE-2026-44463 | High | 2026-05-28 | Not applicable | Proof of concept only | Zed is a code editor. Prior to 0.229.0, Zed's terminal tool permission system can be bypassed by prepending environment variable assignments to allowlisted commands, hijacking program behavior (e.g., PAGER) to execute arbitrary code. This vulnerability is fixed in 0.229.0. |
| CVE-2026-44462 | High | 2026-05-28 | Not applicable | Proof of concept only | Zed is a code editor. Prior to 0.229.0, Zed's terminal tool permission system can be bypassed via bash variable expansion chaining (${var@P}), allowing arbitrary command execution under an allowlisted command prefix. This vulnerability is fixed in 0.229.0. |
| CVE-2026-45037 | High | 2026-05-15 | Not applicable | No public exploit | Tabby (formerly Terminus) is a highly configurable terminal emulator. Prior to 1.0.232, Tabby's terminal linkifier passes any detected URI directly to the operating system's protocol handler without validating the protocol scheme. This allows a malicious SSH or Telnet server to send crafted terminal output containing dangerous protocol URIs which Tabby renders as clickable links, triggering arbitrary OS protocol handlers on the victim's machine. This vulnerability is fixed in 1.0.232. |
| CVE-2026-42590 | High | 2026-05-14 | Not applicable | Proof of concept only | Gotenberg is a Docker-powered stateless API for PDF files. Prior to 8.30.0, The ExifTool metadata write blocklist in Gotenberg can be bypassed using ExifTool's group-prefix syntax, enabling arbitrary file rename, move, hardlink, and symlink creation on the server. ExifTool supports group-prefix syntax where File:FileName is processed identically to FileName -- the prefix is stripped by SetNewValue in Writer.pl before tag matching. The safeKeyPattern regex (^[a-zA-Z0-9\-_.:]+$) allows colons, so prefixed tag names pass validation. Any prefix works: File:FileName, System:Directory, a:HardLink, etc. Additionally, FilePermissions, FileUserID, and FileGroupID pseudo-tags are not blocked at all and can modify file attributes without any prefix. This vulnerability is fixed in 8.30.0. |
| CVE-2026-40893 | High | 2026-05-14 | Not applicable | Proof of concept only | Gotenberg is a Docker-powered stateless API for PDF files. Prior to 8.31.0, Gotenberg only checks if the tag is exactly FileName, so System:FileName slips right through and ExifTool happily renames the file. This allows remote attackers to move, rename, and change permissions for arbitrary files. This vulnerability is fixed in 8.31.0. |
| CVE-2026-43991 | High | 2026-05-12 | Not applicable | No public exploit | JunoClaw is an agentic AI platform built on Juno Network. Prior to 0.x.y-security-1, substring-based blocklist in plugin-shell's command-safety check could be bypassed by adversarial argument constructions, allowing unauthorized command execution on the host when combined with the companion advisory. Pre-patch, the check was applied to the raw command string rather than the parsed first token. This vulnerability is fixed in 0.x.y-security-1. |
| CVE-2026-43929 | High | 2026-05-12 | Not applicable | Proof of concept only | ssrfcheck is a library that checks if a string contains a potential SSRF attack. In 1.3.0 and earlier, ssrfcheck fails to block Server-Side Request Forgery attacks when the target private IP address is encoded as an IPv4-mapped IPv6 address (e.g. http://[::ffff:127.0.0.1]/). The WHATWG URL parser built into Node.js silently normalizes the IPv4 notation inside the brackets to compressed hex form ([::ffff:7f00:1]) before the library's private-IP regex ever runs. The regex was written to match dot-notation only and therefore never matches any real input — all seven IANA private IPv4 ranges, including the AWS/GCP/Azure metadata address 169.254.169.254, are bypassed. Any application using isSSRFSafeURL() to guard HTTP requests made with user-supplied URLs is fully exposed to SSRF. |
| CVE-2026-45006 | High | 2026-05-11 | Not applicable | No public exploit | OpenClaw before 2026.4.23 contains an improper access control vulnerability in the gateway tool's config.apply and config.patch operations that allows compromised models to write unsafe configuration changes by bypassing an incomplete denylist protection. Attackers can persist malicious config modifications affecting command execution, network behavior, credentials, and operator policies that survive restart. |
| CVE-2026-44993 | Low | 2026-05-11 | Not applicable | No public exploit | OpenClaw before 2026.4.20 contains a message classification vulnerability in Feishu card-action callbacks that misclassifies direct messages as group conversations. Attackers can bypass dmPolicy enforcement by triggering card-action flows in direct message conversations that should have been blocked by restrictive policies. |
| CVE-2026-44115 | High | 2026-05-06 | Not applicable | No public exploit | OpenClaw before 2026.4.22 contains an exec allowlist analysis vulnerability allowing shell expansion hiding in unquoted heredoc bodies. Attackers can bypass allowlist validation by embedding shell expansion tokens in heredoc bodies to execute unapproved commands at runtime. |
| CVE-2026-44114 | High | 2026-05-06 | Not applicable | No public exploit | OpenClaw before 2026.4.20 fails to properly reserve the OPENCLAW_ runtime-control environment namespace in workspace dotenv files, allowing attackers to override critical runtime variables. Malicious workspaces can set variables like OPENCLAW_GIT_DIR to manipulate trusted OpenClaw runtime behavior during source-update or installer flows. |
| CVE-2026-43584 | High | 2026-05-06 | Not applicable | No public exploit | OpenClaw before 2026.4.10 contains an insufficient environment variable denylist vulnerability in its exec environment policy that allows operator-supplied overrides of high-risk interpreter startup variables including VIMINIT, EXINIT, LUA_INIT, and HOSTALIASES. Attackers can exploit this by manipulating these environment variables to influence downstream execution behavior or network connectivity. |
| CVE-2026-43578 | Critical | 2026-05-06 | Not applicable | No public exploit | OpenClaw versions 2026.3.31 before 2026.4.10 contain a privilege escalation vulnerability where heartbeat owner downgrade detection misses local background async exec completion events. Attackers can exploit this by providing untrusted completion content to leave a run in a more privileged context than intended. |
| CVE-2026-41934 | High | 2026-05-06 | Not applicable | No public exploit | Vvveb before version 1.0.8.2 contains an authenticated remote code execution vulnerability in the admin code editor that allows low-privilege authenticated users to execute arbitrary code through insufficient file extension restrictions, with the uploaded payload then executable via subsequent unauthenticated HTTP requests. Attackers with editor, author, contributor, or site_admin roles can write a malicious .htaccess file to map arbitrary extensions to the PHP handler, then upload PHP code with that extension to achieve unauthenticated remote code execution when the file is accessed via HTTP. |
| CVE-2026-43566 | Critical | 2026-05-05 | Not applicable | No public exploit | OpenClaw versions 2026.4.7 before 2026.4.14 contain a privilege escalation vulnerability where heartbeat owner downgrade logic skips webhook wake events carrying untrusted content. Attackers can exploit this by sending untrusted webhook wake events to preserve owner-like execution context when the run should have been downgraded. |
| CVE-2026-43532 | Medium | 2026-05-05 | Not applicable | No public exploit | OpenClaw versions 2026.4.7 before 2026.4.10 fail to normalize Discord event cover image parameters in sandbox media processing. Attackers can bypass media normalization to inject host-local media references into channel action paths expecting normalized media. |
| CVE-2026-42435 | High | 2026-05-05 | Not applicable | No public exploit | OpenClaw versions from 2026.2.22 before 2026.4.12 contain an insufficient shell-wrapper detection vulnerability allowing attackers to inject environment variable assignments at the argv level. Attackers can bypass exec preflight handling to manipulate high-risk shell variables like SHELLOPTS and PS4, affecting execution semantics and security controls. |
| CVE-2026-42427 | Medium | 2026-04-28 | Not applicable | No public exploit | OpenClaw before 2026.4.8 contains a remote code execution vulnerability caused by missing environment variable denylist entries for HGRCPATH, CARGO_BUILD_RUSTC_WRAPPER, RUSTC_WRAPPER, and MAKEFLAGS. Attackers can inject malicious build tool environment variables to influence host exec commands and achieve arbitrary code execution. |
| CVE-2026-41915 | Medium | 2026-04-28 | Not applicable | No public exploit | OpenClaw before 2026.4.8 fails to remove git plumbing environment variables from the execution environment before host exec operations. Attackers can exploit this by setting GIT_DIR and related variables to redirect git operations and compromise repository integrity. |
| CVE-2026-41392 | Medium | 2026-04-28 | Not applicable | No public exploit | OpenClaw before 2026.3.31 contains an exec allowlist bypass vulnerability allowing attackers to inherit allowlist trust via shell init-file wrapper invocations. Attackers can exploit shell options like --rcfile, --init-file, and --startup-file to load attacker-chosen initialization files while bypassing exec allowlist matching restrictions. |
| CVE-2026-41391 | Medium | 2026-04-28 | Not applicable | No public exploit | OpenClaw before 2026.3.31 fails to properly sanitize PIP_INDEX_URL and UV_INDEX_URL environment variables in host execution contexts, allowing attackers to redirect Python package-index traffic. Attackers can exploit this bypass to intercept or manipulate package management operations by injecting malicious index URLs through unsanitized environment variables. |
| CVE-2026-31952 | High | 2026-04-24 | Not applicable | No public exploit | Xibo is an open source digital signage platform with a web content management system and Windows display player software. Versions 1.7 through 4.4.0 have an SQL injection vulnerability in the API routes inside the CMS responsible for Filtering DataSets. This allows an authenticated user to to obtain and modify arbitrary data from the Xibo database by injecting specially crafted values in to the API filter parameter. Exploitation of the vulnerability is possible on behalf of an authorized user who has either of the `Access to DataSet Feature` privilege or the `Access to the Layout Feature` privilege. Users should upgrade to version 4.4.1 which fixes this issue. Customers who host their CMS with Xibo Signage have been patched if they are using 4.4, 4.3, 3.3, 2.3 or 1.8. Upgrading to a fixed version is necessary to remediate. Patches are available for earlier versions of Xibo CMS that are out of support, namely 3.3, 2.3, and 1.8. |
| CVE-2026-41361 | Medium | 2026-04-23 | Not applicable | No public exploit | OpenClaw before 2026.3.28 contains an SSRF guard bypass vulnerability that fails to block four IPv6 special-use ranges. Attackers can exploit this by crafting URLs targeting internal or non-routable IPv6 addresses to bypass SSRF protections. |
| CVE-2026-41332 | Medium | 2026-04-23 | Not applicable | No public exploit | OpenClaw before 2026.3.28 contains an environment variable sanitization vulnerability where GIT_TEMPLATE_DIR and AWS_CONFIG_FILE are not blocked in the host-env blocklist. Attackers can exploit approved exec requests to redirect git or AWS CLI behavior through attacker-controlled configuration files to execute untrusted code or load malicious credentials. |
| CVE-2026-41264 | Critical | 2026-04-23 | Not applicable | Working exploit published | Flowise is a drag & drop user interface to build a customized large language model flow. Prior to 3.1.0, the specific flaw exists within the run method of the CSV_Agents class. The issue results from the lack of proper sandboxing when evaluating an LLM generated python script. An attacker can leverage this vulnerability to execute code in the context of the user running the server. Using prompt injection techniques, an unauthenticated attacker with the ability to send prompts to a chatflow using the CSV Agent node may convince an LLM to respond with a malicious python script that executes attacker controlled commands on the Flowise server. This vulnerability is fixed in 3.1.0. |
| CVE-2026-41206 | Medium | 2026-04-23 | Not applicable | Proof of concept only | PySpector is a static analysis security testing (SAST) Framework engineered for modern Python development workflows. The plugin security validator in PySpector uses AST-based static analysis to prevent dangerous code from being loaded as plugins. Prior to version 0.1.8, the blocklist implemented in `PluginSecurity.validate_plugin_code` is incomplete and can be bypassed using several Python constructs that are not checked. An attacker who can supply a plugin file can achieve arbitrary code execution within the PySpector process when that plugin is installed and executed. Version 0.1.8 fixes the issue. |
| CVE-2026-34415 | Critical | 2026-04-22 | Not applicable | Working exploit published | Xerte Online Toolkits versions 3.15 and earlier contain an incomplete input validation vulnerability in the elFinder connector endpoint that fails to block PHP-executable extensions .php4 due to an incorrect regex pattern. Unauthenticated attackers can exploit this flaw combined with authentication bypass and path traversal vulnerabilities to upload malicious PHP code, rename it with a .php4 extension, and execute arbitrary operating system commands on the server. |
| CVE-2026-26274 | Medium | 2026-04-21 | Not applicable | No public exploit | October is a Content Management System (CMS) and web platform. Prior to 3.7.14 and 4.1.10, a vulnerability was identified in the Twig sandbox security policy that allowed database write operations when cms.safe_mode is enabled. Backend users with Developer permissions could use Twig template markup to execute insert, update, and delete operations on any database table through the query builder, which is included in the sandbox allow-list. This vulnerability is fixed in 3.7.14 and 4.1.10. |
| CVE-2026-26067 | Medium | 2026-04-21 | Not applicable | No public exploit | October is a Content Management System (CMS) and web platform. Prior to 3.7.14 and 4.1.10, a server-side information disclosure vulnerability was identified in the handling of CSS preprocessor files. Backend users with Editor permissions could craft .less, .sass, or .scss files that leverage the compiler's import functionality to read arbitrary files from the server. This worked even with cms.safe_mode enabled. This vulnerability is fixed in 3.7.14 and 4.1.10. |
| CVE-2026-25525 | Medium | 2026-04-20 | Not applicable | Proof of concept only | Magento Long Term Support (LTS) is an unofficial, community-driven project provides an alternative to the Magento Community Edition e-commerce platform with a high level of backward compatibility. Prior to version 20.17.0, the Dataflow module in OpenMage LTS uses a weak blacklist filter (`str_replace('../', '', $input)`) to prevent path traversal attacks. This filter can be bypassed using patterns like `..././` or `....//`, which after the replacement still result in `../`. An authenticated administrator can exploit this to read arbitrary files from the server filesystem. Version 20.17.0 patches the issue. |
| CVE-2026-40077 | Low | 2026-04-09 | Not applicable | Proof of concept only | Beszel is a server monitoring platform. Prior to 0.18.7, some API endpoints in the Beszel hub accept a user-supplied system ID and proceed without further checks that the user should have access to that system. As a result, any authenticated user can access these routes for any system if they know the system's ID. System IDs are random 15 character alphanumeric strings, and are not exposed to all users. However, it is theoretically possible for an authenticated user to enumerate a valid system ID via web API. To use the containers endpoints, the user would also need to enumerate a container ID, which is 12 digit hexadecimal string. This vulnerability is fixed in 0.18.7. |
| CVE-2026-39315 | Medium | 2026-04-09 | Not applicable | Proof of concept only | Unhead is a document head and template manager. Prior to 2.1.13, useHeadSafe() is the composable that Nuxt's own documentation explicitly recommends for rendering user-supplied content in <head> safely. Internally, the hasDangerousProtocol() function in packages/unhead/src/plugins/safe.ts decodes HTML entities before checking for blocked URI schemes (javascript:, data:, vbscript:). The decoder uses two regular expressions with fixed-width digit caps. The HTML5 specification imposes no limit on leading zeros in numeric character references. When a padded entity exceeds the regex digit cap, the decoder silently skips it. The undecoded string is then passed to startsWith('javascript:'), which does not match. makeTagSafe() writes the raw value directly into SSR HTML output. The browser's HTML parser decodes the padded entity natively and constructs the blocked URI. This vulnerability is fixed in 2.1.13. |
| CVE-2026-34177 | Critical | 2026-04-09 | Not applicable | Proof of concept only | Canonical LXD versions 4.12 through 6.7 contain an incomplete denylist in isVMLowLevelOptionForbidden (lxd/project/limits/permissions.go), which omits raw.apparmor and raw.qemu.conf from the set of keys blocked under the restricted.virtual-machines.lowlevel=block project restriction. A remote attacker with can_edit permission on a VM instance in a restricted project can inject an AppArmor rule and a QEMU chardev configuration that bridges the LXD Unix socket into the guest VM, enabling privilege escalation to LXD cluster administrator and subsequently to host root. |
| CVE-2026-35410 | Medium | 2026-04-06 | Not applicable | No public exploit | Directus is a real-time API and App dashboard for managing SQL database content. Prior to 11.16.1, an open redirect vulnerability exists in the login redirection logic. The isLoginRedirectAllowed function fails to correctly identify certain malformed URLs as external, allowing attackers to bypass redirect allow-list validation and redirect users to arbitrary external domains upon successful authentication. This vulnerability is fixed in 11.16.1. |
| CVE-2026-34426 | Medium | 2026-04-02 | Not applicable | Proof of concept only | OpenClaw versions prior to commit b57b680 contain an approval bypass vulnerability due to inconsistent environment variable normalization between approval and execution paths, allowing attackers to inject attacker-controlled environment variables into execution without approval system validation. Attackers can exploit differing normalization logic to discard non-portable keys during approval processing while accepting them at execution time, bypassing operator review and potentially influencing runtime behavior including execution of attacker-controlled binaries. |
| CVE-2026-34425 | Medium | 2026-04-02 | Not applicable | No public exploit | OpenClaw versions prior to commit 8aceaf5 contain a preflight validation bypass vulnerability in shell-bleed protection that allows attackers to execute blocked script content by using piped or complex command forms that the parser fails to recognize. Attackers can craft commands such as piped execution, command substitution, or subshell invocation to bypass the validateScriptFileForShellBleed() validation checks and execute arbitrary script content that would otherwise be blocked. |
| CVE-2026-35000 | High | 2026-04-01 | Not applicable | No public exploit | ChangeDetection.io versions prior to 0.54.7 contain a protection bypass vulnerability in the SafeXPath3Parser implementation that allows attackers to read arbitrary local files by using unblocked XPath 3.0/3.1 functions such as json-doc() and similar file-access primitives. Attackers can exploit the incomplete blocklist of dangerous XPath functions to access sensitive data from the local filesystem. |
| CVE-2026-34430 | High | 2026-04-01 | Not applicable | No public exploit | ByteDance DeerFlow versions prior to commit 92c7a20 contain a sandbox escape vulnerability in bash tool handling that allows attackers to execute arbitrary commands on the host system by bypassing regex-based validation using shell features such as directory changes and relative paths. Attackers can exploit the incomplete shell semantics modeling to read and modify files outside the sandbox boundary and achieve arbitrary command execution through subprocess invocation with shell interpretation enabled. |
| CVE-2026-33628 | Medium | 2026-03-26 | Not applicable | Proof of concept only | Invoice Ninja is a source-available invoice, quote, project and time-tracking app built with Laravel. Invoice line item descriptions in Invoice Ninja v5.13.0 bypass the XSS denylist filter, allowing stored XSS payloads to execute when invoices are rendered in the PDF preview or client portal. The line item description field was not passed through `purify::clean()` before rendering. This is fixed in v5.13.4 by the vendor by adding `purify::clean()` to sanitize line item descriptions. |
| CVE-2026-33396 | Critical | 2026-03-26 | Not applicable | Proof of concept only | OneUptime is an open-source monitoring and observability platform. Prior to version 10.0.35, a low-privileged authenticated user (ProjectMember) can achieve remote command execution on the Probe container/host by abusing Synthetic Monitor Playwright script execution. Synthetic monitor code is executed in VMRunner.runCodeInNodeVM with a live Playwright page object in context. The sandbox relies on a denylist of blocked properties/methods, but it is incomplete. Specifically, _browserType and launchServer are not blocked, so attacker code can traverse `page.context().browser()._browserType.launchServer(...)` and spawn arbitrary processes. Version 10.0.35 contains a patch. |
| CVE-2026-4509 | Low | 2026-03-21 | Not applicable | Proof of concept only | A security flaw has been discovered in PbootCMS up to 3.2.12. This affects an unknown function of the file core/function/file.php of the component File Upload. The manipulation of the argument black results in incomplete blacklist. The attack may be launched remotely. The exploit has been released to the public and may be used for attacks. |
| CVE-2026-33139 | High | 2026-03-20 | Not applicable | Proof of concept only | PySpector is a static analysis security testing (SAST) Framework engineered for modern Python development workflows. PySpector versions 0.1.6 and prior are affected by a security validation bypass in the plugin system. The validate_plugin_code() function in plugin_system.py, performs static AST analysis to block dangerous API calls before a plugin is trusted and executed. However, the internal resolve_name() helper only handles ast.Name and ast.Attribute node types, returning None for all others. When a plugin uses indirect function calls via getattr() (such as getattr(os, 'system')) the outer call's func node is of type ast.Call, causing resolve_name() to return None, and the security check to be silently skipped. The plugin incorrectly passes the trust workflow, and executes arbitrary system commands on the user's machine when loaded. This issue has been patched in version 0.1.7. |
| CVE-2026-32940 | Critical | 2026-03-20 | Not applicable | Proof of concept only | SiYuan is a personal knowledge management system. In versions 3.6.0 and below, SanitizeSVG has an incomplete blocklist — it blocks data:text/html and data:image/svg+xml in href attributes but misses data:text/xml and data:application/xml, both of which can render SVG with JavaScript execution. The unauthenticated /api/icon/getDynamicIcon endpoint serves user-controlled input (via the content parameter) directly into SVG markup using fmt.Sprintf with no escaping, served as Content-Type: image/svg+xml. This creates a click-through XSS: a victim navigates to a crafted URL, sees an SVG with an injected link, and clicking it triggers JavaScript via the bypassed MIME types. The attack requires direct navigation to the endpoint or <object>/<embed> embedding, since <img> tag rendering in the frontend doesn't allow interactive links. This issue has been fixed in version 3.6.1. |
| CVE-2026-32747 | Medium | 2026-03-19 | Not applicable | Proof of concept only | SiYuan is a personal knowledge management system. In versions 3.6.0 and below, the globalCopyFiles API eads source files using filepath.Abs() with no workspace boundary check, relying solely on util.IsSensitivePath() whose blocklist omits /proc/, /run/secrets/, and home directory dotfiles. An admin can copy /proc/1/environ or Docker secrets into the workspace and read them via the standard file API. An admin can exfiltrate any file readable by the SiYuan process that falls outside the incomplete blocklist. In containerized deployments this includes all injected secrets and environment variables - a common pattern for passing credentials to containers. The exfiltrated files are then accessible via the standard workspace file API and persist until manually deleted. This issue has been fixed in version 3.6.1. |
| CVE-2026-32022 | Medium | 2026-03-19 | Not applicable | No public exploit | OpenClaw versions prior to 2026.2.21 contain a stdin-only policy bypass vulnerability in the grep tool within tools.exec.safeBins that allows attackers to read arbitrary files by supplying a pattern via the -e flag parameter. Attackers can include a positional filename operand to bypass file access restrictions and read sensitive files.env from the working directory. |
| CVE-2026-32017 | Medium | 2026-03-19 | Not applicable | No public exploit | OpenClaw versions prior to 2026.2.19 contain an allowlist bypass vulnerability in the exec safeBins policy that allows attackers to write arbitrary files using short-option payloads. Attackers can bypass argument validation by attaching short options like -o to whitelisted binaries, enabling unauthorized file-write operations that should be denied by safeBins checks. |
| CVE-2026-31993 | Medium | 2026-03-19 | Not applicable | No public exploit | OpenClaw versions prior to 2026.2.22 contain an allowlist parsing mismatch vulnerability in the macOS companion app that allows authenticated operators to bypass exec approval checks. Attackers with operator.write privileges and a paired macOS beta node can craft shell-chain payloads that pass incomplete allowlist validation and execute arbitrary commands on the paired host. |
| CVE-2026-31992 | High | 2026-03-19 | Not applicable | No public exploit | OpenClaw versions prior to 2026.2.23 contain an allowlist bypass vulnerability in system.run guardrails that allows authenticated operators to execute unintended commands. When /usr/bin/env is allowlisted, attackers can use env -S to bypass policy analysis and execute shell wrapper payloads at runtime. |
| CVE-2026-22175 | High | 2026-03-18 | Not applicable | No public exploit | OpenClaw versions prior to 2026.2.23 contain an exec approval bypass vulnerability in allowlist mode where allow-always grants could be circumvented through unrecognized multiplexer shell wrappers like busybox and toybox sh -c commands. Attackers can exploit this by invoking arbitrary payloads under the same multiplexer wrapper to satisfy stored allowlist rules, bypassing intended execution restrictions. |
| CVE-2026-32128 | Medium | 2026-03-11 | Not applicable | Proof of concept only | FastGPT is an AI Agent building platform. In 4.14.7 and earlier, FastGPT's Python Sandbox (fastgpt-sandbox) includes guardrails intended to prevent file writes (static detection + seccomp). These guardrails are bypassable by remapping stdout (fd 1) to an arbitrary writable file descriptor using fcntl. After remapping, writing via sys.stdout.write() still satisfies the seccomp rule write(fd==1), enabling arbitrary file creation/overwrite inside the sandbox container despite the intended no file writes restriction. |
| CVE-2026-28783 | Critical | 2026-03-04 | Not applicable | No public exploit | Craft is a content management system (CMS). Prior to 5.9.0-beta.1 and 4.17.0-beta.1, Craft CMS implements a blocklist to prevent potentially dangerous PHP functions from being called via Twig non-Closure arrow functions. In order to be able to successfully execute this attack, you need to either have allowAdminChanges enabled on production, or a compromised admin account, or an account with access to the System Messages utility. Several PHP functions are not included in the blocklist, which could allow malicious actors with the required permissions to execute various types of payloads, including RCEs, arbitrary file reads, SSRFs, and SSTIs. This vulnerability is fixed in 5.9.0-beta.1 and 4.17.0-beta.1. |
| CVE-2026-28363 | Critical | 2026-02-27 | Not applicable | No public exploit | In OpenClaw before 2026.2.23, tools.exec.safeBins validation for sort could be bypassed via GNU long-option abbreviations (such as --compress-prog) in allowlist mode, leading to approval-free execution paths that were intended to require approval. Only an exact string such as --compress-program was denied. |
| CVE-2026-1773 | High | 2026-02-24 | Not applicable | No public exploit | IEC 60870-5-104 used in RTU500: Potential Denial of Service impact on reception of invalid U-format frame. Product is only affected if IEC 60870-5-104 bi-directional functionality is configured. Enabling secure communication following IEC 62351-3 does not remediate the vulnerability but mitigates the risk of exploitation. |