Advisories

domain="path" authorization is checked before final file open/use. A symlink swap between check-time and use-time bypasses policy-denied read/write.

domain="path" authorization is checked before final file open/use. A symlink swap between check-time and use-time bypasses policy-denied read/write.

domain="path" authorization is checked before final file open/use. A symlink swap between check-time and use-time bypasses policy-denied read/write.

domain="path" authorization is checked before final file open/use. A symlink swap between check-time and use-time bypasses policy-denied read/write.

domain="path" authorization is checked before final file open/use. A symlink swap between check-time and use-time bypasses policy-denied read/write.

domain="path" authorization is checked before final file open/use. A symlink swap between check-time and use-time bypasses policy-denied read/write.

domain="path" authorization is checked before final file open/use. A symlink swap between check-time and use-time bypasses policy-denied read/write.

domain="path" authorization is checked before final file open/use. A symlink swap between check-time and use-time bypasses policy-denied read/write.

domain="path" authorization is checked before final file open/use. A symlink swap between check-time and use-time bypasses policy-denied read/write.

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

In MAT decoder uses 32-bit arithmetic due to incorrect parenthesization resulting in a heap over-read. ================================================================= ==969652==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x506000003b40 at pc 0x555557b2a926 bp 0x7fffffff4c80 sp 0x7fffffff4c70 READ of size 8 at 0x506000003b40 thread T0

When the flarum/nicknames extension is enabled, a registered user can set their nickname to a string that email clients interpret as a hyperlink. The nickname is inserted verbatim into plain-text notification emails, and recipients may be misled into visiting attacker-controlled domains.

A denial of service vulnerability exists in the ASF (WMV/WMA) file type detection parser. When parsing a crafted input where an ASF sub-header has a size field of zero, the parser enters an infinite loop. The payload value becomes negative (-24), causing tokenizer.ignore(payload) to move the read position backwards, so the same sub-header is read repeatedly forever. Any application that uses file-type to detect the type of untrusted/attacker-controlled input is …

An unauthenticated attacker can send a crafted GET request directly to /oauth/:provider/callback with a forged profile in the query string. The OAuth service's authentication payload has a fallback chain that reaches params.query (the raw request query) when Grant's session/state responses are empty. Since the attacker never initiated an OAuth authorize flow, Grant has no session to work with and produces no response, so the fallback fires. The forged profile then …

Socket.IO clients can send arbitrary JavaScript objects as the id argument to any service method (get, patch, update, remove). The transport layer performs no type checking on this argument. When the service uses the MongoDB adapter, these objects pass through getObjectId() and land directly in the MongoDB query as operators. Sending {$ne: null} as the id matches every document in the collection.

At the rate limit filter, if we enabled the response phase limit with apply_on_stream_done in the rate limit configuration and the response phase limit request fails directly, it may crash Envoy.

Note: This vulnerability was originally reported to the Google OSS VRP (Issue ID: 477542544). The Google Security Team requested that I coordinate directly with the Envoy maintainers for triage and remediation. I am submitting this report here to facilitate that process. Technical Details I have identified a logic vulnerability in Envoy's HTTP connection manager (FilterManager) that allows for Zombie Stream Filter Execution. This issue creates a "Use-After-Free" (UAF) or state-corruption …

Calling Utility::getAddressWithPort with a scoped IPv6 addresses causes a crash. This utility is called in the data plane from the original_src filter and the dns filter.

The Envoy RBAC (Role-Based Access Control) filter contains a logic vulnerability in how it validates HTTP headers when multiple values are present for the same header name. Instead of validating each header value individually, Envoy concatenates all values into a single comma-separated string. This behavior allows attackers to bypass RBAC policies—specifically "Deny" rules—by sending duplicate headers, effectively obscuring the malicious value from exact-match mechanisms.

An off-by-one write in Envoy::JsonEscaper::escapeString() can corrupt std::string null-termination, causing undefined behavior and potentially leading to crashes or out-of-bounds reads when the resulting string is later treated as a C-string.

t.String({ format: 'url' }) is vulnerable to redos Repeating a partial url format (protocol and hostname) multiple times cause regex to slow down significantly 'http://a'.repeat(n) Here's a table demonstrating how long it takes to process repeated partial url format | n repeat | elapsed_ms | | — | — | | 1024 | 33.993 | | 2048 | 134.357 | | 4096 | 537.608 | | 8192 | 2155.842 | …

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-crjq-wm6x-6qx7. This link is maintained to preserve external references. Original Description Incorrect default permissions in .NET allows an authorized attacker to elevate privileges locally.

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-crjq-wm6x-6qx7. This link is maintained to preserve external references. Original Description Incorrect default permissions in .NET allows an authorized attacker to elevate privileges locally.

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-crjq-wm6x-6qx7. This link is maintained to preserve external references. Original Description Incorrect default permissions in .NET allows an authorized attacker to elevate privileges locally.

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-crjq-wm6x-6qx7. This link is maintained to preserve external references. Original Description Incorrect default permissions in .NET allows an authorized attacker to elevate privileges locally.

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-crjq-wm6x-6qx7. This link is maintained to preserve external references. Original Description Incorrect default permissions in .NET allows an authorized attacker to elevate privileges locally.

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-crjq-wm6x-6qx7. This link is maintained to preserve external references. Original Description Incorrect default permissions in .NET allows an authorized attacker to elevate privileges locally.

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-73j8-2gch-69rq. This link is maintained to preserve external references. Original Description Out-of-bounds read in .NET allows an unauthorized attacker to deny service over a network.

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-4vgm-c2wm-63mw. This link is maintained to preserve external references. Original Description Allocation of resources without limits or throttling in ASP.NET Core allows an unauthorized attacker to deny service over a network.

An Insecure Direct Object Reference (IDOR) vulnerability exists in Craft Commerce’s cart functionality that allows users to hijack any shopping cart by knowing or guessing its 32-character number. This vulnerability enables the takeover of shopping sessions and potential exposure of PII.

A stored XSS vulnerability exists when a user tries to update the Order Status from the Commerce Orders Table. The Order Status Name is rendered without proper escaping, allowing script execution to occur.

Craft Commerce is vulnerable to SQL Injection in the purchasables table endpoint. The sort parameter is split by | and the first part (column name) is passed directly as an array key to orderBy() without whitelist validation. Yii2's query builder does NOT escape array keys, allowing an authenticated attacker to inject arbitrary SQL into the ORDER BY clause.

Craft Commerce is vulnerable to SQL Injection in the inventory levels table data endpoint. The sort[0][direction] and sort[0][sortField] parameters are concatenated directly into an addOrderBy() clause without any validation or sanitization. An authenticated attacker with access to the Commerce Inventory section can inject arbitrary SQL queries, potentially leading to a full database compromise.

A stored XSS vulnerability exists in the Commerce Settings - Inventory Locations page. The Name field is rendered without proper HTML escaping, allowing an attacker to execute arbitrary JavaScript. This XSS triggers when an administrator (or user with product editing permissions) creates or edits a variant product.

A Stored Cross-Site Scripting (XSS) vulnerability exists in the Craft Commerce Order details. Malicious JavaScript can be injected via the Shipping Method Name, Order Reference, or Site Name. When a user opens the order details slideout via a double-click on the order index page, the injected payload executes.

Stored XSS vulnerabilities exist in the Commerce Inventory page. The Product Title, Variant Title, and Variant SKU fields are rendered without proper HTML escaping, allowing an attacker to execute arbitrary JavaScript when any user (including administrators) views the inventory management page. This vulnerability enables session hijacking by fetching the PHP Info utility page, which displays unmasked session cookies. Unlike other XSS chains that require elevated sessions, this attack provides instant …

Craft CMS has a CSRF issue in the preview token endpoint at /actions/preview/create-token. The endpoint accepts an attacker-supplied previewToken. Because the action does not require POST and does not enforce a CSRF token, an attacker can force a logged-in victim editor to mint a preview token chosen by the attacker. That token can then be used by the attacker (without authentication) to access previewed/unpublished content tied to the victim’s authorized …

The nohtml config option, intended to prevent execution of JavaScript in user-uploaded HTML files, did not apply to SVG images.

Camaleon CMS versions 2.4.5.0 through 2.9.1, prior to commit f54a77e, contain a path traversal vulnerability in the AWS S3 uploader implementation that allows authenticated users to read arbitrary files from the web server’s filesystem. The issue occurs in the download_private_file functionality when the application is configured to use the CamaleonCmsAwsUploader backend. Unlike the local uploader implementation, the AWS uploader does not validate file paths with valid_folder_path?, allowing directory traversal sequences …

Server-Side Request Forgery (SSRF) in Azure MCP Server allows an authorized attacker to elevate privileges over a network.

Server-Side Request Forgery (SSRF) in Azure MCP Server allows an authorized attacker to elevate privileges over a network.

Server-Side Request Forgery (SSRF) in Azure MCP Server allows an authorized attacker to elevate privileges over a network.

This issue affects the ExtractEmbeddedFiles example in Apache PDFBox: from 2.0.24 through 2.0.35, from 3.0.0 through 3.0.6. The ExtractEmbeddedFiles example contains a path traversal vulnerability (CWE-22) because the filename that is obtained from PDComplexFileSpecification.getFilename() is appended to the extraction path. Users who have copied this example into their production code should review it to ensure that the extraction path is acceptable. The example has been changed accordingly, now the initial …

Actual Sync Server allows authenticated users to upload files through POST /sync/upload-user-file. In versions prior to 26.3.0, improper validation of the user-controlled x-actual-file-id header means that traversal segments (../) can escape the intended directory and write files outside userFiles.

The /api/user/git-config endpoint constructs shell commands by interpolating user-supplied gitName and gitEmail values into command strings passed to child_process.exec(). The input is placed within double quotes and only " is escaped, but backticks (`), $() command substitution, and \ sequences are all interpreted within double-quoted strings in bash. This allows authenticated attackers to execute arbitrary OS commands via the git configuration endpoint.

The SSRF protection fix for https://github.com/vllm-project/vllm/security/advisories/GHSA-qh4c-xf7m-gxfc can be bypassed in the load_from_url_async method due to inconsistent URL parsing behavior between the validation layer and the actual HTTP client.

A privilege escalation vulnerability exists in the publish service of SiYuan Note that allows a low-privilege publish account (RoleReader) to modify notebook content via the /api/block/appendHeadingChildren API endpoint. The endpoint only requires model.CheckAuth, which accepts RoleReader sessions. Because the endpoint performs a persistent document mutation and does not enforce CheckAdminRole or CheckReadonly, a publish user with read-only privileges can append new blocks to existing documents. This allows remote authenticated publish …

The OIDC token endpoint rejects an authorization code only when both the client ID is wrong and the code is expired. This allows cross-client code exchange and expired code reuse.

A flaw in callback URL validation allowed crafted redirect_uri values containing URL userinfo (@) to bypass legitimate callback pattern checks. If an attacker can trick a user into opening a malicious authorization link, the authorization code may be redirected to an attacker-controlled host.

The Google, Apple, and Facebook authentication adapters use JWT verification to validate identity tokens. When the adapter's audience configuration option is not set (clientId for Google/Apple, appIds for Facebook), JWT verification silently skips audience claim validation. This allows an attacker to use a validly signed JWT issued for a different application to authenticate as any user on the target Parse Server. For Google and Apple, the vulnerability is exploitable when …

When graphQLPublicIntrospection is disabled, __type queries nested inside inline fragments (e.g. … on Query { __type(name:"User") { name } }) bypass the introspection control, allowing unauthenticated users to perform type reconnaissance. __schema introspection is not affected.

The file metadata endpoint (GET /files/:appId/metadata/:filename) does not enforce beforeFind / afterFind file triggers. When these triggers are used as access-control gates, the metadata endpoint bypasses them entirely, allowing unauthorized access to file metadata. This affects any deployment that relies on Parse.Cloud.beforeFind(Parse.File, …) to restrict file access. Only file metadata (user-defined key-value pairs set via addMetadata) is exposed; file content remains protected.

The PagesRouter static file serving route is vulnerable to a path traversal attack that allows unauthenticated reading of files outside the configured pagesPath directory. The boundary check uses a string prefix comparison without enforcing a directory separator boundary. An attacker can use path traversal sequences to access files in sibling directories whose names share the same prefix as the pages directory (e.g. pages-secret starts with pages). This affects any Parse …

OpenClaw's system.run shell-wrapper detection did not recognize PowerShell -EncodedCommand forms as inline-command wrappers. In allowlist mode, a caller with access to system.run could invoke pwsh or powershell using -EncodedCommand, -enc, or -e, and the request would fall back to plain argv analysis instead of the normal shell-wrapper approval path. This could allow a PowerShell inline payload to execute without the approval step that equivalent -Command invocations would require. Latest published …

OpenClaw's hooks HTTP handler counted hook authentication failures before rejecting unsupported HTTP methods. An unauthenticated client could send repeated non-POST requests (for example GET) with an invalid token to consume the hook auth failure budget and trigger the temporary lockout window for that client key. The fix moves the hook method gate ahead of auth-failure accounting so unsupported methods return 405 Method Not Allowed without incrementing the hook auth limiter.

OpenClaw's macOS Dashboard flow exposed Gateway authentication material to browser-controlled surfaces. Before the fix, the macOS app appended the shared Gateway token and password to the Dashboard URL query string when opening the Control UI in the browser. The Control UI then imported the token and persisted it into browser localStorage under openclaw.control.settings.v1. This expanded exposure of reusable Gateway admin credentials into browser address-bar/query surfaces and persistent script-readable storage.

system.run env override sanitization allowed dangerous override-only helper-command pivots to reach subprocesses. A caller who could invoke system.run with env overrides could bypass allowlist/approval intent by steering an allowlisted tool through helper-command or config-loading environment variables such as GIT_SSH_COMMAND, editor/pager hooks, and GIT_CONFIG_* / NPM_CONFIG_*.

OpenClaw's system.run dispatch-wrapper handling applied different depth-boundary rules to shell-wrapper approval detection and execution planning. With exactly four transparent dispatch wrappers such as repeated env invocations before /bin/sh -c, the approval classifier could stop treating the command as a shell wrapper at the depth boundary while execution planning still unwrapped through to the shell payload. In security=allowlist mode, that mismatch could skip the expected approval-required path for the shell wrapper …

OpenClaw's system.run dispatch-wrapper handling applied different depth-boundary rules to shell-wrapper approval detection and execution planning. With exactly four transparent dispatch wrappers such as repeated env invocations before /bin/sh -c, the approval classifier could stop treating the command as a shell wrapper at the depth boundary while execution planning still unwrapped through to the shell payload. In security=allowlist mode, that mismatch could skip the expected approval-required path for the shell wrapper …

OpenClaw's system.run allowlist analysis did not honor POSIX shell comment semantics when deriving allow-always persistence entries. A caller in security=allowlist mode who received an allow-always decision could submit a shell command whose tail was commented out at runtime, for example by using an unquoted # before a chained payload. The runtime shell would execute only the pre-comment portion, but allowlist persistence could still analyze and store the non-executed tail as …

Sandboxed requester sessions could reach host-side ACP session initialization through /acp spawn. OpenClaw already blocked sessions_spawn({ runtime: "acp" }) from sandboxed sessions, but the slash-command path initialized ACP directly without applying the same host-runtime guard first.

Sandboxed requester sessions could reach host-side ACP session initialization through /acp spawn. OpenClaw already blocked sessions_spawn({ runtime: "acp" }) from sandboxed sessions, but the slash-command path initialized ACP directly without applying the same host-runtime guard first.

OpenClaw's fetchWithSsrFGuard(…) followed cross-origin redirects while preserving arbitrary caller-supplied headers except for a narrow denylist (Authorization, Proxy-Authorization, Cookie, Cookie2). This allowed custom authorization headers such as X-Api-Key, Private-Token, and similar sensitive headers to be forwarded to a different origin after a redirect. The fix switches cross-origin redirect handling from a narrow sensitive-header denylist to a safe-header allowlist, so only benign headers such as content negotiation and cache validators survive an …

OpenClaw's fetchWithSsrFGuard(…) followed cross-origin redirects while preserving arbitrary caller-supplied headers except for a narrow denylist (Authorization, Proxy-Authorization, Cookie, Cookie2). This allowed custom authorization headers such as X-Api-Key, Private-Token, and similar sensitive headers to be forwarded to a different origin after a redirect. The fix switches cross-origin redirect handling from a narrow sensitive-header denylist to a safe-header allowlist, so only benign headers such as content negotiation and cache validators survive an …

/allowlist … –store resolved the selected channel accountId for reads, but store writes still dropped that accountId and wrote into the legacy unscoped pairing allowlist store. Because default-account reads still merge legacy unscoped entries, a store entry intended for one account could silently authorize the same sender on the default account. This is a real cross-account sender-authorization scoping bug. Severity is set to medium because exploitation requires an already-authorized user …

A gateway client authenticated with operator.write could route /config set or /config unset through chat.send and reach persistent config mutation even though direct config RPC methods are admin-scoped.

OneUptime's GitHub App callback trusts attacker-controlled state and installation_id values and updates Project.gitHubAppInstallationId with isRoot: true without validating that the caller is authorized for the target project. This allows an attacker to overwrite another project's GitHub App installation binding. Related GitHub endpoints also lack effective authorization, so a valid installation ID can be used to enumerate repositories and create CodeRepository records in an arbitrary project.

A vulnerability in the filestring() function of the nltk.util module in nltk version 3.9.2 allows arbitrary file read due to improper validation of input paths. The function directly opens files specified by user input without sanitization, enabling attackers to access sensitive system files by providing absolute paths or traversal paths. This vulnerability can be exploited locally or remotely, particularly in scenarios where the function is used in web APIs or …

A user assigned the platform-user role can retrieve WireGuard private keys of all wireguard configs in a network by calling GET /api/extclients/{network} or GET /api/nodes/{network}. While the Netmaker UI restricts visibility, the API endpoints return full records, including private keys, without filtering based on the requesting user's ownership. Credits Artem Danilov (Positive Technologies)

The user update handler (PUT /api/users/{username}) lacks validation to prevent an admin-role user from assigning the super-admin role during account updates. While the code correctly blocks an admin from assigning the admin role to another user, it does not include an equivalent check for the super-admin role. Credits Artem Danilov (Positive Technologies)

The Authorise middleware in Netmaker incorrectly validates host JWT tokens. When a route permits host authentication (hostAllowed=true), a valid host token bypasses all subsequent authorisation checks without verifying that the host is authorised to access the specific requested resource. Any entity possessing knowledge of object identifiers (node IDs, host IDs) can craft a request with an arbitrary valid host token to access, modify, or delete resources belonging to other hosts. …

Kubewarden is a policy engine for Kubernetes. Kubewarden cluster operators can grant permissions to users to deploy namespaced AdmissionPolicies and AdmissionPolicyGroups in their Namespaces. One of Kubewarden promises is that configured users can deploy namespaced policies in a safe manner, without privilege escalation. An attacker with privileged "AdmissionPolicy" create permissions (which isn't the default) could make use of 3 deprecated host-callback APIs: kubernetes/ingresses, kubernetes/namespaces, kubernetes/services. The attacker can craft a …

The TimescaleDB export module constructs SQL queries using string concatenation with unsanitized system monitoring data. The normalize() method wraps string values in single quotes but does not escape embedded single quotes, making SQL injection trivial via attacker-controlled data such as process names, filesystem mount points, network interface names, or container names. Root Cause: The normalize() function uses f"'{value}'" for string values without escaping single quotes within the value. The resulting …

The /api/4/config REST API endpoint returns the entire parsed Glances configuration file (glances.conf) via self.config.as_dict() with no filtering of sensitive values. The configuration file contains credentials for all configured backend services including database passwords, API tokens, JWT signing keys, and SSL key passwords.

Stored XSS is possible via share metadata fields (e.g., title, description) that are rendered into HTML for /public/share/<hash> without context-aware escaping. The server uses text/template instead of html/template, allowing injected scripts to execute when victims visit the share URL.

The remediation for CVE-2026-27611 appears incomplete. Password protected shares still disclose tokenized downloadURL via /public/api/share/info in docker image gtstef/filebrowser:1.3.1-webdav-2.

AzuraCast's ConfigWriter::cleanUpString() method fails to sanitize Liquidsoap string interpolation sequences (#{…}), allowing authenticated users with StationPermissions::Media or StationPermissions::Profile permissions to inject arbitrary Liquidsoap code into the generated configuration file. When the station is restarted and Liquidsoap parses the config, #{…} expressions are evaluated, enabling arbitrary command execution via Liquidsoap's process.run() function.

A vulnerability in Apache IoTDB. This issue affects Apache IoTDB: from 1.0.0 before 1.3.7, from 2.0.0 before 2.0.7. Users are recommended to upgrade to version 1.3.7 or 2.0.7, which fixes the issue.

Improper Input Validation vulnerability in Apache IoTDB. This issue affects Apache IoTDB: from 1.0.0 before 1.3.7, from 2.0.0 before 2.0.7. Users are recommended to upgrade to version 1.3.7 or 2.0.7, which fixes the issue.

A user with access to the DB could craft a database entry that would result in executing code on Triggerer - which gives anyone who have access to DB the same permissions as Dag Author. Since direct DB access is not usual and recommended for Airflow, the likelihood of it making any damage is low. Users should upgrade to version 6.0.0 of the provider to avoid even that risk.

In AWS Auth manager, the origin of the SAML authentication has been used as provided by the client and not verified against the actual instance URL. This allowed to gain access to different instances with potentially different access controls by reusing SAML response from other instances. You should upgrade to 9.22.0 version of provider if you use AWS Auth Manager.

Register unwilling users for events (potential harassment/spam) Cancel other users' event participation Manipulate event participant counts and comments If events have participation limits, fill slots with unwanted registrations

Location: packages/server/src/integrations/postgres.ts:529-531

A security vulnerability exists in outdated versions of the x402 SDK. This vulnerability does not affect users' private keys, smart contracts, or funds. The issue impacts resource servers accepting payments on Solana when the facilitator is running a vulnerable version of the x402 SDK.

A security vulnerability exists in outdated versions of the x402 SDK. This vulnerability does not affect users' private keys, smart contracts, or funds. The issue impacts resource servers accepting payments on Solana when the facilitator is running a vulnerable version of the x402 SDK.

A security vulnerability exists in outdated versions of the x402 SDK. This vulnerability does not affect users' private keys, smart contracts, or funds. The issue impacts resource servers accepting payments on Solana when the facilitator is running a vulnerable version of the x402 SDK.

This impacts users of Shescape that configure their shell to point to a file on disk that is a link to a link. The precise result of being affected depends on the actual shell used and incorrect shell identified by Shescape. In particular, an attacker may be able to bypass escaping for the shell being used. This can result, for example, in exposure of sensitive information, consider the following proof …

A critical unauthenticated remote code execution (RCE) vulnerability exists in the MCP stdio configuration validation introduced in version 2.0.5. The application allows unrestricted user registration, meaning any attacker can create an account and exploit the command injection flaw. Despite implementing a whitelist for allowed commands (npx, uvx) and blacklists for dangerous arguments and environment variables, the validation can be bypassed using the -p flag with npx node. This allows any …

Soroban host ensures that MuxedAddress objects can't be used as storage keys in order to proactively prevent the contract logic bugs. However, due to a bug in Soroban host implementation, a failure in Val->ScVal conversion during the storage key computation will have the flag indicating that storage conversion is happening stuck in the true state until the next storage access. While the flag is stuck in true state, any MuxedAddress …

A path traversal vulnerability in the /export endpoint allows an attacker to read arbitrary files from the server filesystem. By exploiting double‑encoded traversal sequences, an attacker can access sensitive files such as conf/conf.json, which contains secrets including the API token, cookie signing key, and workspace access authentication code. Leaking these secrets may enable administrative access to the SiYuan kernel API, and in certain deployment scenarios could potentially be chained into …

A path traversal vulnerability in the /export endpoint allows an attacker to read arbitrary files from the server filesystem. By exploiting double‑encoded traversal sequences, an attacker can access sensitive files such as conf/conf.json, which contains secrets including the API token, cookie signing key, and workspace access authentication code. Leaking these secrets may enable administrative access to the SiYuan kernel API, and in certain deployment scenarios could potentially be chained into …

This impacts users of Shescape that configure their shell to point to a file on disk that is a link to a link. The precise result of being affected depends on the actual shell used and incorrect shell identified by Shescape. In particular, an attacker may be able to bypass escaping for the shell being used. This can result, for example, in exposure of sensitive information, consider the following proof …

In version 1.20.0, when using new sync streams with config.edition: 3, certain subquery filters were ignored when determining which data to sync to users. Depending on the sync stream configuration, this could result in authenticated users syncing data that should have been restricted. Only queries that gate synchronization using subqueries without partitioning the result set are affected. Not affected: Sync rules (bucket_definitions) Sync streams using config.edition: 2 No data is …

In version 1.20.0, when using new sync streams with config.edition: 3, certain subquery filters were ignored when determining which data to sync to users. Depending on the sync stream configuration, this could result in authenticated users syncing data that should have been restricted. Only queries that gate synchronization using subqueries without partitioning the result set are affected. Not affected: Sync rules (bucket_definitions) Sync streams using config.edition: 2 No data is …

In version 1.20.0, when using new sync streams with config.edition: 3, certain subquery filters were ignored when determining which data to sync to users. Depending on the sync stream configuration, this could result in authenticated users syncing data that should have been restricted. Only queries that gate synchronization using subqueries without partitioning the result set are affected. Not affected: Sync rules (bucket_definitions) Sync streams using config.edition: 2 No data is …

In version 1.20.0, when using new sync streams with config.edition: 3, certain subquery filters were ignored when determining which data to sync to users. Depending on the sync stream configuration, this could result in authenticated users syncing data that should have been restricted. Only queries that gate synchronization using subqueries without partitioning the result set are affected. Not affected: Sync rules (bucket_definitions) Sync streams using config.edition: 2 No data is …

OneUptime allows project members to run custom Playwright/JavaScript code via Synthetic Monitors to test websites. However, the system executes this untrusted user code inside the insecure Node.js vm module. By leveraging a standard prototype-chain escape (this.constructor.constructor), an attacker can bypass the sandbox, gain access to the underlying Node.js process object, and execute arbitrary system commands (RCE) on the oneuptime-probe container. Furthermore, because the probe holds database/cluster credentials in its environment …

OneUptime allows project members to run custom Playwright/JavaScript code via Synthetic Monitors to test websites. However, the system executes this untrusted user code inside the insecure Node.js vm module. By leveraging a standard prototype-chain escape (this.constructor.constructor), an attacker can bypass the sandbox, gain access to the underlying Node.js process object, and execute arbitrary system commands (RCE) on the oneuptime-probe container. Furthermore, because the probe holds database/cluster credentials in its environment …

Summary OneUptime Synthetic Monitors allow low-privileged project users to submit custom Playwright code that is executed on the oneuptime-probe service. In the current implementation, this untrusted code is run inside Node's vm and is given live host Playwright objects such as browser and page. This creates a distinct server-side RCE primitive: the attacker does not need the classic this.constructor.constructor(…) sandbox escape. Instead, the attacker can directly use the injected Playwright …

Summary OneUptime Synthetic Monitors allow low-privileged project users to submit custom Playwright code that is executed on the oneuptime-probe service. In the current implementation, this untrusted code is run inside Node's vm and is given live host Playwright objects such as browser and page. This creates a distinct server-side RCE primitive: the attacker does not need the classic this.constructor.constructor(…) sandbox escape. Instead, the attacker can directly use the injected Playwright …

The Meta Box plugin for WordPress is vulnerable to arbitrary file deletion due to insufficient file path validation in the 'ajax_delete_file' function in all versions up to, and including, 5.11.1. This makes it possible for authenticated attackers, with Contributor-level access and above, to delete arbitrary files on the server, which can easily lead to remote code execution when the right file is deleted (such as wp-config.php).

When the HTTP server is enabled (MCP_HTTP_ENABLED=true), the application configures FastAPI's CORSMiddleware with allow_origins=[''], allow_credentials=True, allow_methods=[""], and allow_headers=["*"]. The wildcard Access-Control-Allow-Origin: * header permits any website to read API responses cross-origin. When combined with anonymous access (MCP_ALLOW_ANONYMOUS_ACCESS=true) - the simplest way to get the HTTP dashboard working without OAuth - no credentials are needed, so any malicious website can silently read, modify, and delete all stored memories.

When the HTTP server is enabled (MCP_HTTP_ENABLED=true), the application configures FastAPI's CORSMiddleware with allow_origins=[''], allow_credentials=True, allow_methods=[""], and allow_headers=["*"]. The wildcard Access-Control-Allow-Origin: * header permits any website to read API responses cross-origin. When combined with anonymous access (MCP_ALLOW_ANONYMOUS_ACCESS=true) - the simplest way to get the HTTP dashboard working without OAuth - no credentials are needed, so any malicious website can silently read, modify, and delete all stored memories.

FUXA used a static fallback JWT signing secret (frangoteam751) when no secretCode was configured. If authentication was enabled without explicitly setting a custom secret, an attacker who knew the default value could forge valid JWT tokens and bypass authentication. This issue has been addressed in version 1.3.0 by removing the static fallback and generating a secure random secret when no secretCode is provided.

The User management API endpoints (GET /api/v1/users and GET /api/v1/users/{id}) are accessible to any authenticated user without admin/owner role verification, exposing all users' email addresses, roles, and account status.

The /objects/playlistsFromUser.json.php endpoint returns all playlists for any user without requiring authentication or authorization. An unauthenticated attacker can enumerate user IDs and retrieve playlist information including playlist names, video IDs, and playlist status for any user on the platform.

The /objects/playlistsFromUser.json.php endpoint returns all playlists for any user without requiring authentication or authorization. An unauthenticated attacker can enumerate user IDs and retrieve playlist information including playlist names, video IDs, and playlist status for any user on the platform.

Hostname verification in Apache ZooKeeper ZKTrustManager falls back to reverse DNS (PTR) when IP SAN validation fails, allowing attackers who control or spoof PTR records to impersonate ZooKeeper servers or clients with a valid certificate for the PTR name. It's important to note that attacker must present a certificate which is trusted by ZKTrustManager which makes the attack vector harder to exploit. Users are recommended to upgrade to version 3.8.6 …

Improper handling of configuration values in ZKConfig in Apache ZooKeeper 3.8.5 and 3.9.4 on all platforms allows an attacker to expose sensitive information stored in client configuration in the client's logfile. Configuration values are exposed at INFO level logging rendering potential production systems affected by the issue. Users are recommended to upgrade to version 3.8.6 or 3.9.5 which fixes this issue.

A path traversal vulnerability in archive extraction allows a specifically crafted Zarf package to create symlinks pointing outside the destination directory, enabling arbitrary file read or write on the system processing the package.

A vulnerability involving tool name collision and indirect prompt injection allows a malicious remote MCP server to hijack tool execution. By exploiting an ambiguous naming convention in the MCP client (mcp_{service}_{tool}), an attacker can register a malicious tool that overwrites a legitimate one (e.g., tavily_extract). This enables the attacker to redirect LLM execution flow, exfiltrate system prompts, context, and potentially execute other tools with the user's privileges.

A critical Remote Code Execution (RCE) vulnerability exists in the application's database query functionality. The validation system fails to recursively inspect child nodes within PostgreSQL array expressions and row expressions, allowing attackers to bypass SQL injection protections. By smuggling dangerous PostgreSQL functions inside these expressions and chaining them with large object operations and library loading capabilities, an unauthenticated attacker can achieve arbitrary code execution on the database server with database …

An authorization bypass in tenant management endpoints of WeKnora application allows any authenticated user to read, modify, or delete any tenant by ID. Since account registration is open to the public, this vulnerability allows any unauthenticated attacker to register an account and subsequently exploit the system. This enables cross-tenant account takeover and destruction, making the impact critical.

A cross-tenant authorization bypass in the knowledge base copy endpoint allows any authenticated user to clone (duplicate) another tenant’s knowledge base into their own tenant by knowing/guessing the source knowledge base ID. This enables bulk data exfiltration (document/FAQ content) across tenants, making the impact critical.

A DNS rebinding vulnerability in the web_fetch tool allows an unauthenticated attacker to bypass URL validation and access internal resources on the server, including private IP addresses (e.g., 127.0.0.1, 192.168.x.x). By crafting a malicious domain that resolves to a public IP during validation and subsequently resolves to a private IP during execution, an attacker can access sensitive local services and potentially exfiltrate data.

A broken access control vulnerability in the database query tool allows any authenticated tenant to read sensitive data belonging to other tenants, including API keys, model configurations, and private messages. The application fails to enforce tenant isolation on critical tables (models, messages, embeddings), enabling unauthorized cross-tenant data access with user-level authentication privileges.

createWebhook() in Vercel Workflow DevKit accepts a user-specified token parameter that serves as the credential for the public webhook endpoint /.well-known/workflow/v1/webhook/{token}. Official documentation recommended predictable token patterns, making it possible for an unauthenticated remote attacker to guess the token and inject arbitrary payloads into the workflow execution context.

createWebhook() in Vercel Workflow DevKit accepts a user-specified token parameter that serves as the credential for the public webhook endpoint /.well-known/workflow/v1/webhook/{token}. Official documentation recommended predictable token patterns, making it possible for an unauthenticated remote attacker to guess the token and inject arbitrary payloads into the workflow execution context.

An authenticated SSH user can force the server to make HTTP requests to internal/private IP addresses by running repo import with a crafted –lfs-endpoint URL. The initial batch request is blind (the response from a metadata endpoint won't parse as valid LFS JSON), but an attacker hosting a fake LFS server can chain this into full read access to internal services by returning download URLs that point at internal targets.

Snipe-IT versions prior to 8.3.7 contain sensitive user attributes related to account privileges that are insufficiently protected against mass assignment. An authenticated, low-privileged user can craft a malicious API request to modify restricted fields of another user account, including the Super Admin account. By changing the email address of the Super Admin and triggering a password reset, an attacker can fully take over the Super Admin account, resulting in complete …

A Server-Side Request Forgery (SSRF) vulnerability in the /download endpoint allows any user with API access to induce the PinchTab server to make requests to arbitrary URLs, including internal network services and local system files, and exfiltrate the full response content.

The readOnlyMasterKey can be used to create and delete files via the Files API (POST /files/:filename, DELETE /files/:filename). This bypasses the read-only restriction which violates the access scope of the readOnlyMasterKey. Any Parse Server deployment that uses readOnlyMasterKey and exposes the Files API is affected. An attacker with access to the readOnlyMasterKey can upload arbitrary files or delete existing files.

The readOnlyMasterKey can call POST /loginAs to obtain a valid session token for any user. This allows a read-only credential to impersonate arbitrary users with full read and write access to their data. Any Parse Server deployment that uses readOnlyMasterKey is affected.

A malformed $regex query parameter (e.g. [abc) causes the database to return a structured error object that is passed unsanitized through the API response. This leaks database internals such as error messages, error codes, code names, cluster timestamps, and topology details. The vulnerability is exploitable by any client that can send query requests, depending on the deployment's permission configuration.

Mercurius fails to enforce the configured queryDepth limit on GraphQL subscription queries received over WebSocket connections. The depth check is correctly applied to HTTP queries and mutations, but subscription queries are parsed and executed without invoking the depth validation. This allows a remote client to submit arbitrarily deeply nested subscription queries over WebSocket, bypassing the intended depth restriction. On schemas with recursive types, this can lead to denial of service …

A security vulnerability has been identified in GitHub Copilot CLI's shell tool that could allow arbitrary code execution through crafted bash parameter expansion patterns. An attacker who can influence the commands executed by the agent (e.g., via prompt injection through repository files, MCP server responses, or user instructions) can exploit bash parameter transformation operators to execute hidden commands, bypassing the safety assessment that classifies commands as "read-only."

The NVIDIA NIM router (/api/v1/nvidia-nim/*) is whitelisted in the global authentication middleware, allowing unauthenticated access to privileged container management and token generation endpoints.

The Flowise platform has a critical Insecure Direct Object Reference (IDOR) vulnerability combined with a Business Logic Flaw in the PUT /api/v1/loginmethod endpoint. While the endpoint requires authentication, it fails to validate if the authenticated user has ownership or administrative rights over the target organizationId. This allows any low-privileged user (including "Free" plan users) to: Overwrite the SSO configuration of any other organization. Enable "Enterprise-only" features (SSO/SAML) without a license. …

Flowise trusts any HTTP client that sets the header x-request-from: internal, allowing an authenticated tenant session to bypass all /api/v1/** authorization checks. With only a browser cookie, a low-privilege tenant can invoke internal administration endpoints (API key management, credential stores, custom function execution, etc.), effectively escalating privileges.

1. Root Cause The vulnerability stems from relying solely on the MIME type without cross-validating the file extension or actual content. This allows attackers to upload executable files (e.g., .js, .php) or malicious scripts (.html) by masquerading them as benign images or documents. 2. Key Attack Scenarios Server Compromise (RCE): An attacker uploads a Web Shell and triggers its execution on the server. Successful exploitation grants system privileges, allowing unauthorized …

A Mass Assignment vulnerability in the /api/v1/leads endpoint allows any unauthenticated user to control internal entity fields (id, createdDate, chatId) by including them in the request body. The endpoint uses Object.assign() to copy all properties from the request body to the Lead entity without any input validation or field filtering. This allows attackers to bypass auto-generated fields and inject arbitrary values. | Field | Value | |——-|——-| | Vulnerability Type …

The default keyGenerator in express-rate-limit applies IPv6 subnet masking (/56 by default) to all addresses that net.isIPv6() returns true for. This includes IPv4-mapped IPv6 addresses (::ffff:x.x.x.x), which Node.js returns as request.ip on dual-stack servers. Because the first 80 bits of all IPv4-mapped addresses are zero, a /56 (or any /32 to /80) subnet mask produces the same network key (::/56) for every IPv4 client. This collapses all IPv4 traffic into …

The _findContentBySchemaText method in src/defuddle.ts interpolates image src and alt attributes directly into an HTML string without escaping: html += <img src="${imageSrc}" alt="${imageAlt}">; An attacker can use a " in the alt attribute to break out of the attribute context and inject event handlers. This is a separate vulnerability from the sanitization bypass fixed in f154cb7 — the injection happens during string construction, not in the DOM, so _stripUnsafeElements cannot …

A Denial of Service vulnerability exists in CoreDNS's loop detection plugin that allows an attacker to crash the DNS server by sending specially crafted DNS queries. The vulnerability stems from the use of a predictable pseudo-random number generator (PRNG) for generating a secret query name, combined with a fatal error handler that terminates the entire process.

A logical vulnerability in CoreDNS allows DNS access controls to be bypassed due to the default execution order of plugins. Security plugins such as acl are evaluated before the rewrite plugin, resulting in a Time-of-Check Time-of-Use (TOCTOU) flaw.

The DisallowedRawHtml extension can be bypassed by inserting a newline, tab, or other ASCII whitespace character between a disallowed HTML tag name and the closing >. For example, <script\n> would pass through unfiltered and be rendered as a valid HTML tag by browsers. This is a cross-site scripting (XSS) vector for any application that relies on this extension to sanitize untrusted user input. All applications using the DisallowedRawHtml extension to …

The vars_regexp matcher in vars.go:337 double-expands user-controlled input through the Caddy replacer. When vars_regexp matches against a placeholder like {http.request.header.X-Input}, the header value gets resolved once (expected), then passed through repl.ReplaceAll() again (the bug). This means an attacker can put {env.DATABASE_URL} or {file./etc/passwd} in a request header and the server will evaluate it, leaking environment variables, file contents, and system info. header_regexp does NOT do this — it passes header …

Caddy's forward_auth directive with copy_headers generates conditional header-set operations that only fire when the upstream auth service includes the named header in its response. No delete or remove operation is generated for the original client-supplied request header with the same name. When an auth service returns 200 OK without one of the configured copy_headers headers, the client-supplied header passes through unchanged to the backend. Any requester holding a valid authentication …

zeptoclaw implements a allowlist combined with a blocklist to prevent malicious shell commands in src/security/shell.rs. However, even in the Strict mode, attackers can completely bypass all the guards from allowlist and blocklist: to bypass the allowlist, command injection is enough, such as ;, $() etc. to bypass the REGEX_BLOCKED_PATTERNS, argument injection is enough, such as the python3 -P -c "…" to bypass the LITERAL_BLOCKED_PATTERNS, file name wildcards can do the …

zeptoclaw implements a blocklist to prevent dangerous commands running in android device shell, but this blocklist has several blocked commands with argements in the pattern literal, such as rm -f and rm -rf, this can be simply bypassed by using different orders for these arguments, such as rm -r -f or rm -fr etc.

The multi-level nested syntax caused a segmentation fault (core dump).

The application's "Import document via URL" feature is vulnerable to Server-Side Request Forgery (SSRF) through HTTP redirects. While the backend implements comprehensive URL validation (blocking private IPs, loopback addresses, reserved hostnames, and cloud metadata endpoints), it fails to validate redirect targets. An attacker can bypass all protections by using a redirect chain, forcing the server to access internal services. Additionally, Docker-specific internal addresses like host.docker.internal are not blocked.

The leak causes the JVM to crash with OOME.

This is a configuration bypass vulnerability that enables arbitrary code execution. The @backstage/plugin-techdocs-node package uses an allowlist to filter dangerous MkDocs configuration keys during the documentation build process. A gap in this allowlist allows attackers to craft an mkdocs.yml that causes arbitrary Python code execution, completely bypassing TechDocs' security controls.

tar (npm) can be tricked into creating a hardlink that points outside the extraction directory by using a drive-relative link target such as C:../target.txt, which enables file overwrite outside cwd during normal tar.x() extraction.

tar (npm) can be tricked into creating a hardlink that points outside the extraction directory by using a drive-relative link target such as C:../target.txt, which enables file overwrite outside cwd during normal tar.x() extraction.

StringM::from_str does not validate that the input length is within the declared maximum (MAX). Calling StringM::<N>::from_str(s) where s is longer than N bytes succeeds and returns an Ok value instead of Err(Error::LengthExceedsMax), producing a StringM that violates its length invariant. This affects any code that constructs StringM values from string input using FromStr (including str::parse), and relies on the type's maximum length constraint being enforced. An oversized StringM could propagate …

StringM::from_str does not validate that the input length is within the declared maximum (MAX). Calling StringM::<N>::from_str(s) where s is longer than N bytes succeeds and returns an Ok value instead of Err(Error::LengthExceedsMax), producing a StringM that violates its length invariant. This affects any code that constructs StringM values from string input using FromStr (including str::parse), and relies on the type's maximum length constraint being enforced. An oversized StringM could propagate …

A vulnerability exists in the Sliver C2 server's Protobuf unmarshalling logic due to a systemic lack of nil-pointer validation. By extracting valid implant credentials and omitting nested fields in a signed message, an authenticated actor can trigger an unhandled runtime panic. Because the mTLS, WireGuard, and DNS transport layers lack the panic recovery middleware present in the HTTP transport, this results in a global process termination. While requiring post-authentication access …

This advisory addresses the use of the search_hub() function within the SageMaker Python SDK's JumpStart search functionality. An actor with the ability to control query parameters passed to the search_hub() function could potentially provide malformed input that causes the eval() function to execute arbitrary commands, access sensitive data, or compromise the execution environment. A defense-in-depth enhancement has been implemented to replace code evaluation with safe string operations when processing search …

An Arbitrary File Read vulnerability exists in the ImageTextPromptValue class in Exploding Gradients RAGAS v0.2.3 to v0.2.14. The vulnerability stems from improper validation and sanitization of URLs supplied in the retrieved_contexts parameter when handling multimodal inputs.

Python-Markdown version 3.8 contain a vulnerability where malformed HTML-like sequences can cause html.parser.HTMLParser to raise an unhandled AssertionError during Markdown parsing. Because Python-Markdown does not catch this exception, any application that processes attacker-controlled Markdown may crash. This enables remote, unauthenticated Denial of Service in web applications, documentation systems, CI/CD pipelines, and any service that renders untrusted Markdown. The issue was acknowledged by the vendor and fixed in version 3.8.1. This …

The edit_package() function implements insufficient sanitization for the pack_folder parameter. The current protection relies on a single-pass string replacement of "../", which can be bypassed using crafted recursive traversal sequences. Exploitation An authenticated user with MODIFY permission can bypass the sanitization by submitting a payload such as: pack_folder=…/./…/./…/./tmp After the single-pass replacement, this becomes: ../../../tmp Because the traversal sequences are not properly validated, the resulting normalized path escapes the intended …

A security vulnerability exists in the Plane project management platform that allows unauthenticated attackers to enumerate workspace members and extract sensitive information including email addresses, user roles, and internal identifiers. The vulnerability stems from Django REST Framework permission classes being incorrectly configured to allow anonymous access to protected endpoints. This vulnerability enables attackers to: Enumerate all members of any workspace without authentication Extract user email addresses and personally identifiable information …

The webhook URL validation in plane/app/serializers/webhook.py only checks ip.is_loopback, allowing attackers with workspace ADMIN role to create webhooks pointing to private/internal network addresses (10.x.x.x, 172.16.x.x, 192.168.x.x, 169.254.169.254, etc.). When webhook events fire, the server makes requests to these internal addresses and stores the response — enabling SSRF with full response read-back.

Pingora versions prior to 0.8.0 would immediately forward bytes following a request with an Upgrade header to the backend, without waiting for a 101 Switching Protocols response. This allows an attacker to smuggle requests to the backend and bypass proxy-level security controls. This vulnerability primarily affects standalone Pingora deployments where a Pingora proxy is exposed to external traffic. An attacker could exploit this to bypass proxy-level ACL controls and WAF …

Pingora versions prior to 0.8.0 generated cache keys using only the URI path, excluding critical factors such as the host header. This allows an attacker to poison the cache and serve cross-origin responses to users. This vulnerability affects users of Pingora's alpha proxy caching feature who relied on the default CacheKey implementation. An attacker could exploit this for cross-tenant data leakage in multi-tenant deployments, or serve malicious content to legitimate …

Pingora versions prior to 0.8.0 improperly allowed HTTP/1.0 request bodies to be close-delimited and incorrectly handled multiple Transfer-Encoding values. This allows an attacker to desync Pingora's request framing from backend servers and smuggle requests to the backend. This vulnerability primarily affects standalone Pingora deployments in front of certain backends that accept HTTP/1.0 requests. An attacker could exploit this to bypass proxy-level ACL controls and WAF logic, poison caches and upstream …

Parse Server's readOnlyMasterKey option allows access with master-level read privileges but is documented to deny all write operations. However, some endpoints incorrectly accept the readOnlyMasterKey for mutating operations. This allows a caller who only holds the readOnlyMasterKey to create, modify, and delete Cloud Hooks and to start Cloud Jobs, which can be used for data exfiltration. Any Parse Server deployment that uses the readOnlyMasterKey option is affected. Note than an …

pac4j-jwt versions prior to 4.5.9, 5.7.9, and 6.3.3 contain an authentication bypass vulnerability in JwtAuthenticator when processing encrypted JWTs that allows remote attackers to forge authentication tokens. Attackers who possess the server's RSA public key can create a JWE-wrapped PlainJWT with arbitrary subject and role claims, bypassing signature verification to authenticate as any user including administrators.

The Jetty URI parser has some key differences compared to other common parsers when evaluating invalid or unusual URIs. Specifically:

A Server-Side Request Forgery (SSRF) vulnerability was identified in the @opennextjs/cloudflare package, resulting from a path normalization bypass in the /cdn-cgi/image/ handler. The @opennextjs/cloudflare worker template includes a /cdn-cgi/image/ handler intended for development use only. In production, Cloudflare's edge intercepts /cdn-cgi/image/ requests before they reach the Worker. However, by substituting a backslash for a forward slash (/cdn-cgi\image/ instead of /cdn-cgi/image/), an attacker can bypass edge interception and have the request …

An authentication context confusion vulnerability in RestartAction allows a low‑privileged authenticated user to execute actions they are not permitted to run. RestartAction constructs a new internal connect.Request without preserving the original caller’s authentication headers or cookies. When this synthetic request is passed to StartAction, the authentication resolver falls back to the guest user. If the guest account has broader permissions than the authenticated caller, this results in privilege escalation and …

An authentication context confusion vulnerability in RestartAction allows a low‑privileged authenticated user to execute actions they are not permitted to run. RestartAction constructs a new internal connect.Request without preserving the original caller’s authentication headers or cookies. When this synthetic request is passed to StartAction, the authentication resolver falls back to the guest user. If the guest account has broader permissions than the authenticated caller, this results in privilege escalation and …

OliveTin does not revoke server-side sessions when a user logs out. Although the browser cookie is cleared, the corresponding session remains valid in server storage until expiry (default ≈ 1 year). An attacker with a previously stolen or captured session cookie can continue authenticating after logout, resulting in a post-logout authentication bypass. This is a session management flaw that violates expected logout semantics.

OliveTin does not revoke server-side sessions when a user logs out. Although the browser cookie is cleared, the corresponding session remains valid in server storage until expiry (default ≈ 1 year). An attacker with a previously stolen or captured session cookie can continue authenticating after logout, resulting in a post-logout authentication bypass. This is a session management flaw that violates expected logout semantics.

When JWT authentication is configured using either: authJwtPubKeyPath (local RSA public key), or authJwtHmacSecret (HMAC secret), the configured audience value (authJwtAud) is not enforced during token parsing. As a result, validly signed JWT tokens with an incorrect aud claim are accepted for authentication. This allows authentication using tokens intended for a different audience/service.

When JWT authentication is configured using either: authJwtPubKeyPath (local RSA public key), or authJwtHmacSecret (HMAC secret), the configured audience value (authJwtAud) is not enforced during token parsing. As a result, validly signed JWT tokens with an incorrect aud claim are accepted for authentication. This allows authentication using tokens intended for a different audience/service.

An unauthenticated attacker can trigger server-side panics by first creating an execution log entry with a nil binding via StartActionByGet (invalid action ID), then calling KillAction or RestartAction on that tracking ID. This causes a nil-pointer dereference in API handlers and results in repeated per-request panics (Empty reply from server), enabling denial of service through panic/log/CPU amplification.

An authorization flaw in OliveTin allows authenticated users with view: false permission to enumerate action bindings and metadata via dashboard and API endpoints. Although execution (exec) may be correctly denied, the backend does not enforce IsAllowedView() when constructing dashboard and action binding responses. As a result, restricted users can retrieve action titles, IDs, icons, and argument metadata.

The /api/backup endpoint is accessible without authentication and discloses the encryption keys required to decrypt the backup in the X-Backup-Security response header. This allows an unauthenticated attacker to download a full system backup containing sensitive data (user credentials, session tokens, SSL private keys, Nginx configurations) and decrypt it immediately.

A vulnerability in Multer versions < 2.1.1 allows an attacker to trigger a Denial of Service (DoS) by sending malformed requests, potentially causing stack overflow.

A CRLF Injection vulnerability in MimeKit 4.15.0 allows an attacker to embed \r\n into the SMTP envelope address local-part (when the local-part is a quoted-string). This is non-compliant with RFC 5321 and can result in SMTP command injection (e.g., injecting additional RCPT TO / DATA / RSET commands) and/or mail header injection, depending on how the application uses MailKit/MimeKit to construct and send messages. The issue becomes exploitable when the …

A Cross-Site Request Forgery (CSRF) vulnerability was identified in Mercurius versions 16. The issue arises from incorrect parsing of the Content-Type header in requests. Specifically, requests with Content-Type values such as application/x-www-form-urlencoded, multipart/form-data, or text/plain could be misinterpreted as application/json. This misinterpretation bypasses the preflight checks performed by the fetch() API, potentially allowing unauthorized actions to be performed on behalf of an authenticated user.

The /api/health/detailed endpoint returns detailed system information including OS version, Python version, CPU count, memory totals, disk usage, and the full database filesystem path. When MCP_ALLOW_ANONYMOUS_ACCESS=true is set (required for the HTTP server to function without OAuth/API key), this endpoint is accessible without authentication. Combined with the default 0.0.0.0 binding, this exposes sensitive reconnaissance data to the entire network.

Leantime v2.3.27 is vulnerable to Stored HTML Injection. The firstname and lastname fields in the admin user edit page are rendered without HTML escaping, allowing an authenticated user to inject arbitrary HTML that executes when the profile is viewed.

LangGraph checkpointers can load msgpack-encoded checkpoints that reconstruct Python objects during deserialization. If an attacker can modify checkpoint data in the backing store (for example, after a database compromise or other privileged write access to the persistence layer), they can potentially supply a crafted payload that triggers unsafe object reconstruction when the checkpoint is loaded. This is a post-exploitation / defense-in-depth issue. Exploitation requires the ability to write attacker-controlled checkpoint …

A flaw was found in org.keycloak.broker.saml. When a disabled Security Assertion Markup Language (SAML) client is configured as an Identity Provider (IdP)-initiated broker landing target, it can still complete the login process and establish a Single Sign-On (SSO) session. This allows a remote attacker to gain unauthorized access to other enabled clients without re-authentication, effectively bypassing security restrictions. A fix is available at https://github.com/keycloak/keycloak/releases/tag/26.5.5.

A security flaw in the IdentityBrokerService.performLogin endpoint of Keycloak allows authentication to proceed using an Identity Provider (IdP) even after it has been disabled by an administrator. An attacker who knows the IdP alias can reuse a previously generated login request to bypass the administrative restriction. This undermines access control enforcement and may allow unauthorized authentication through a disabled external provider.

If a malicious authenticated user uploads SVG and creates a hotlink for it, they achieve stored XSS.

A registered user without privileges to create or modify file requests is able to create a short-lived API key that has the permission to do so. The user must be registered with Gokapi. If you do not have any other users with access to the admin/upload menu, you are not impacted.

A privilege escalation vulnerability in the user rank demotion logic allows a demoted user's existing API keys to retain ApiPermManageFileRequests and ApiPermManageLogs permissions, enabling continued access to upload-request management and log viewing endpoints after the user has been stripped of all privileges.

The upload status SSE implementation on /uploadStatus publishes global upload state to any authenticated listener and includes file_id values that are not scoped to the requesting user.

The login flow accepts credential-bearing requests without CSRF protection mechanisms tied to the browser session context. The handler parses form values directly and creates a session on successful credential validation. Issue found by aisafe.io

A Stored Cross-site Scripting (XSS) vulnerability exists in the comment and issue description functionality. The application's HTML sanitizer explicitly allows data: URI schemes, enabling authenticated users to inject arbitrary JavaScript execution via malicious links.

Stored XSS is still possible through unsafe template rendering that mixes user input with safe() plus permissive sanitizer handling of data URLs.

There is a security issue in Gogs where deleting a release can fail if a user-controlled tag name is passed to Git without the right separator, allowing Git option injection and therefore interfering with the process.

It was confirmed in a test environment that an attacker can store an HTML/JavaScript payload in a repository’s Milestone name, and when another user selects that Milestone on the New Issue page (/issues/new), a DOM-Based XSS is triggered.

Overwritable LFS object across different repos leads to supply-chain attack, all LFS objects are vulnerable to be maliciously overwritten by malicious attackers.

The Gogs API still accepts tokens in URL parameters such as token and access_token, which can leak through logs, browser history, and referrers.

Incomplete CSRF protections around /session/verify made it possible to use OTCs in login sessions different from the requesting session. In some scenarios this might have made it easier for phishers to take over a Ghost site.

Incomplete CSRF protections around /session/verify made it possible to use OTCs in login sessions different from the requesting session. In some scenarios this might have made it easier for phishers to take over a Ghost site.

Fonoster 0.5.5 before 0.6.1 allows ../ directory traversal to read arbitrary files via the /sounds/:file or /tts/:file VoiceServer endpoint. This occurs in serveFiles in mods/voice/src/utils.ts. NOTE: serveFiles exists in 0.5.5 but not in the next release, 0.6.1.

The /api/v1/account/forgot-password endpoint returns the full user object including PII (id, name, email, status, timestamps) in the response body instead of a generic success message. This exposes sensitive user information to unauthenticated attackers who only need to know a valid email address.

The default bcrypt salt rounds is set to 5, which is below the recommended minimum for security.

Fastify incorrectly accepts malformed Content-Type headers containing trailing characters after the subtype token, in violation of RFC 9110 §8.3.1. For example, a request sent with Content-Type: application/json garbage passes validation and is processed normally, rather than being rejected with 415 Unsupported Media Type. When regex-based content-type parsers are in use (a documented Fastify feature), the malformed value is matched against registered parsers using the full string including the trailing garbage. …

The official example script examples/recursively_extract_attachments.py contains a path traversal vulnerability that allows arbitrary file write outside the intended output directory. Attachment filenames extracted from parsed emails are directly used to construct output file paths without any sanitization, allowing an attacker-controlled filename to escape the target directory.

Vulnerability Allowing MFA Bypass Affected EC-CUBE Versions Versions: 4.1.0 – 4.3.1 Vulnerability Overview If an administrator’s ID and password are compromised, an issue exists that allows an attacker to bypass the normally required two-factor authentication (2FA) and log in to the administrative interface. Severity and Impact CVSS v3.1 score Base score: 6.2 / Temporal score: 5.7 / Environmental score (after mitigation and countermeasures): 0.0 An attacker can forcibly overwrite the …

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-xq2h-p299-vjwv. This link is maintained to preserve external references. Original Description An HTTP request smuggling vulnerability (CWE-444) was found in Pingora's handling of HTTP/1.1 connection upgrades. The issue occurs when a Pingora proxy reads a request containing an Upgrade header, causing the proxy to pass through the rest of the bytes on the connection to a backend …

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-hj7x-879w-vrp7. This link is maintained to preserve external references. Original Description An HTTP Request Smuggling vulnerability (CWE-444) has been found in Pingora's parsing of HTTP/1.0 and Transfer-Encoding requests. The issue occurs due to improperly allowing HTTP/1.0 request bodies to be close-delimited and incorrect handling of multiple Transfer-Encoding values, allowing attackers to send HTTP/1.0 requests in a way …

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-f93w-pcj3-rggc. This link is maintained to preserve external references. Original Description A cache poisoning vulnerability has been found in the Pingora HTTP proxy framework’s default cache key construction. The issue occurs because the default HTTP cache key implementation generates cache keys using only the URI path, excluding critical factors such as the host header (authority). Operators relying …

This issue affects Docker CLI through 29.1.5

This issue affects Docker CLI through 29.1.5

This issue affects Docker CLI through 29.1.5

An open redirect vulnerability exists in django-allauth versions prior to 65.14.1 when SAML IdP initiated SSO is enabled (it is disabled by default), which may allow an attacker to redirect users to an arbitrary external website via a crafted URL.

What kind of vulnerability is it? Who is impacted? A path traversal vulnerability exists in dbt-common's safe_extract() function used when extracting tarball archives. The function uses os.path.commonprefix() to validate that extracted files remain within the intended destination directory. However, commonprefix() compares paths character-by-character rather than by path components, allowing a malicious tarball to write files to sibling directories with matching name prefixes. For example, when extracting to /tmp/packages, a crafted …

What kind of vulnerability is it? Who is impacted? A path traversal vulnerability exists in dbt-common's safe_extract() function used when extracting tarball archives. The function uses os.path.commonprefix() to validate that extracted files remain within the intended destination directory. However, commonprefix() compares paths character-by-character rather than by path components, allowing a malicious tarball to write files to sibling directories with matching name prefixes. For example, when extracting to /tmp/packages, a crafted …

Inappropriate user token revocation due to a logic error in the token revocation endpoint implementation in Cloudfoundry UAA v77.30.0 to v78.7.0 and in Cloudfoundry Deployment v48.7.0 to v54.10.0.

A vulnerability in the SCM URL parsing used by Backstage integrations allowed path traversal sequences in encoded form to be included in file paths. When these URLs were processed by integration functions that construct API URLs, the traversal segments could redirect requests to unintended SCM provider API endpoints using the configured server-side integration credentials. This affects instances that use any of the SCM integrations (GitHub, Bitbucket Server, Bitbucket Cloud) with …

The official docker-compose.yml publishes the memcached service on host port 11211 (0.0.0.0:11211) with no authentication, while the Dockerfile configures PHP to store all user sessions in that memcached instance. An attacker who can reach port 11211 can read, modify, or flush session data — enabling session hijacking, admin impersonation, and mass session destruction without any application-level authentication.

When converting MCP tools/call request to OpenAPI request, input path, query, and header values are not sanitized.

When converting MCP tools/call request to OpenAPI request, input path, query, and header values are not sanitized.

The time-sync crate attempted to exfiltrate .env files to a server that was in turn impersonating the legitimate timeapi.io service. This the same attack that we've seen three times in the last few days. The malicious crate had 1 version published on 2026-03-04 approximately 50 minutes before removal and had no evidence of actual downloads. There were no crates depending on this crate on crates.io.

The dnp3times crate attempted to exfiltrate .env files to a server that was in turn impersonating the legitimate timeapi.io service. It was loosely trying to typosquat the dnp3time crate, but otherwise was the same attack as the recent time_calibrator and time_calibrators malware. The malicious crate had 1 version published on 2026-03-04 approximately 6 hours before removal and had no evidence of actual downloads. There were no crates depending on this …

An Observable Timing Discrepancy in @perfood/couch-auth v0.26.0 allows attackers to access sensitive information via a timing side-channel.

A host header injection vulnerability in the mailer component of @perfood/couch-auth v0.26.0 allows attackers to obtain reset tokens and execute an account takeover via spoofing the HTTP Host header.

A malicious scaffolder template can bypass the log redaction mechanism to exfiltrate secrets provided run through task event logs. The attack requires: The ability to register a template in the catalog A victim who executes the malicious template

A vulnerability in Zitadel's login V2 interface was discovered, allowing for possible account takeover.

A vulnerability in Zitadel's login V2 interface was discovered, allowing for possible account takeover.

A vulnerability in Zitadel's login V2 UI allowed users to bypass login behavior and security policies and self-register new accounts or sign in using password even if corresponding options were disabled in their organizaton.

A vulnerability in Zitadel's login V2 UI allowed users to bypass login behavior and security policies and self-register new accounts or sign in using password even if corresponding options were disabled in their organizaton.

A vulnerability was discovered in Zitadel's login V2 interface that allowed a possible account takeover.

A vulnerability was discovered in Zitadel's login V2 interface that allowed a possible account takeover.

The Blog Application is vulnerable to Stored Cross-Site Scripting (XSS) via the Blog Post Title. The vulnerability arises because the post title is injected directly into the HTML tag without proper escaping. An attacker with permissions to create or edit blog posts can inject malicious JavaScript into the title field. This script will execute in the browser of any user (including administrators) who views the blog post. This leads to …

Testing confirmed that even when a Manager has manage=false for a given collection, they can still perform the following management operations as long as they have access to the collection: PUT /api/organizations/<org_id>/collections/<col_id> succeeds (HTTP 200) PUT /api/organizations/<org_id>/collections/<col_id>/users succeeds (HTTP 200) DELETE /api/organizations/<org_id>/collections/<col_id> succeeds (HTTP 200)

In the test environment, it was confirmed that an authenticated regular user can specify another user’s cipher_id and call: PUT /api/ciphers/{id}/partial Even though the standard retrieval API correctly denies access to that cipher, the partial update endpoint returns 200 OK and exposes cipherDetails (including name, notes, data, secureNote, etc.).

A Manager account (access_all=false) was able to escalate privileges by directly invoking the bulk-access API against collections that were not originally assigned to them. The API allowed changing assigned=false to assigned=true, resulting in unauthorized access. Additionally, prior to the bulk-access call, the regular single-update API correctly returned 401 Unauthorized for the same collection. After executing the bulk-access API, the same update API returned 200 OK, confirming an authorization gap at …

Vaultwarden v1.34.3 and prior are susceptible to a 2FA bypass when performing protected actions. An attacker who gains authenticated access to a user’s account can exploit this bypass to perform protected actions such as accessing the user's API key or deleting the user's vault and organisations the user is an admin/owner of. Note that

There is a potential vulnerability in Traefik managing TLS handshake on TCP routers. When Traefik processes a TLS connection on a TCP router, the read deadline used to bound protocol sniffing is cleared before the TLS handshake is completed. When a TLS handshake read error occurs, the code attempts a second handshake with different connection parameters, silently ignoring the initial error. A remote unauthenticated client can exploit this by sending …

There is a potential vulnerability in Traefik managing TLS handshake on TCP routers. When Traefik processes a TLS connection on a TCP router, the read deadline used to bound protocol sniffing is cleared before the TLS handshake is completed. When a TLS handshake read error occurs, the code attempts a second handshake with different connection parameters, silently ignoring the initial error. A remote unauthenticated client can exploit this by sending …

The ForwardAuth middleware reads the entire authentication server response body into memory using io.ReadAll with no size limit. A single HTTP request through a ForwardAuth-protected route can cause the Traefik process to allocate gigabytes of memory and be killed by the OOM killer, resulting in complete denial of service for all routes on the affected entrypoint.

The ForwardAuth middleware reads the entire authentication server response body into memory using io.ReadAll with no size limit. A single HTTP request through a ForwardAuth-protected route can cause the Traefik process to allocate gigabytes of memory and be killed by the OOM killer, resulting in complete denial of service for all routes on the affected entrypoint.

There is a potential vulnerability in Traefik managing the Connection header with X-Forwarded headers. When Traefik processes HTTP/1.1 requests, the protection put in place to prevent the removal of Traefik-managed X-Forwarded headers (such as X-Real-Ip, X-Forwarded-Host, X-Forwarded-Port, etc.) via the Connection header does not handle case sensitivity correctly. The Connection tokens are compared case-sensitively against the protected header names, but the actual header deletion operates case-insensitively. As a result, a …

There is a potential vulnerability in Traefik managing the Connection header with X-Forwarded headers. When Traefik processes HTTP/1.1 requests, the protection put in place to prevent the removal of Traefik-managed X-Forwarded headers (such as X-Real-Ip, X-Forwarded-Host, X-Forwarded-Port, etc.) via the Connection header does not handle case sensitivity correctly. The Connection tokens are compared case-sensitively against the protected header names, but the actual header deletion operates case-insensitively. As a result, a …

SVGO accepts XML with custom entities, without guards against entity expansion or recursion. This can result in a small XML file (811 bytes) stalling the application and even crashing the Node.js process with JavaScript heap out of memory.

An unauthenticated reflected XSS vulnerability exists in the dynamic icon API endpoint: GET /api/icon/getDynamicIcon When type=8, attacker-controlled content is embedded into SVG output without escaping. Because the endpoint is unauthenticated and returns image/svg+xml, a crafted URL can inject executable SVG/HTML event handlers (for example onerror) and run JavaScript in the SiYuan web origin. This can be chained to perform authenticated API actions and exfiltrate sensitive data when a logged-in user …

In OpenClaw, the sandboxed image tool did not honor tools.fs.workspaceOnly=true for mounted paths resolved by the sandbox FS bridge. This allowed reading out-of-workspace mounted images (for example /agent/*) and forwarding those bytes to vision model providers.

In OpenClaw, the sandboxed image tool did not honor tools.fs.workspaceOnly=true for mounted paths resolved by the sandbox FS bridge. This allowed reading out-of-workspace mounted images (for example /agent/*) and forwarding those bytes to vision model providers.

In certain elevated-mode configurations, tools.elevated.allowFrom accepted broader identity signals than intended. The fix tightens matching to sender-scoped identity by default and makes mutable metadata matching explicit.

In shared Slack workspace deployments that rely on sender restrictions (allowFrom, DM policy, or channel user allowlists), some interactive callbacks (block_action, view_submission, view_closed) could be accepted before full sender authorization checks. In that scenario, an unauthorized workspace member could enqueue system-event text into an active session. This issue did not provide unauthenticated access, cross-gateway isolation bypass, or host-level privilege escalation by itself.

In shared Slack workspace deployments that rely on sender restrictions (allowFrom, DM policy, or channel user allowlists), some interactive callbacks (block_action, view_submission, view_closed) could be accepted before full sender authorization checks. In that scenario, an unauthorized workspace member could enqueue system-event text into an active session. This issue did not provide unauthenticated access, cross-gateway isolation bypass, or host-level privilege escalation by itself.

In certain workspace-restricted configurations, OpenClaw could follow hardlink aliases inside the workspace that reference files outside the workspace boundary. By default, tools.fs.workspaceOnly is off. This primarily affects deployments that intentionally enable workspace-only filesystem restrictions (and workspace-only apply_patch checks).

BlueBubbles is an optional OpenClaw channel plugin. A configuration-sensitive access-control mismatch allowed DM senders to be treated as authorized when dmPolicy was pairing or allowlist and allowFrom was empty/unset.

BlueBubbles is an optional OpenClaw channel plugin. A configuration-sensitive access-control mismatch allowed DM senders to be treated as authorized when dmPolicy was pairing or allowlist and allowFrom was empty/unset.

A malicious or compromised MCP (Model Context Protocol) tool server can exfiltrate arbitrary local files from the host system by injecting MEDIA: directives into tool result text content. OpenClaw's tool result processing pipeline extracts file paths from MEDIA: tokens without source-level validation, passes them through a localRoots allowlist check that includes os.tmpdir() by default (covering /tmp on Linux/macOS and %TEMP% on Windows), and then reads and delivers the file contents …

OpenClaw's SSRF hostname/IP guard did not detect ISATAP embedded IPv4 addresses (…:5efe:w.x.y.z). A crafted URL containing an ISATAP IPv6 literal could embed a private IPv4 target (for example loopback) and bypass private-address filtering in URL-fetching paths.

isPrivateIpv4() in bundled SSRF guard code missed several IPv4 special-use/non-global ranges, so web_fetch could allow targets that should be blocked by SSRF policy.

isPrivateIpv4() in bundled SSRF guard code missed several IPv4 special-use/non-global ranges, so web_fetch could allow targets that should be blocked by SSRF policy.

OpenClaw had account-scope gaps in pairing-store access for DM pairing policy, which could let a pairing approval from one account authorize the same sender on another account in multi-account setups.

OpenClaw had account-scope gaps in pairing-store access for DM pairing policy, which could let a pairing approval from one account authorize the same sender on another account in multi-account setups.

A crafted local avatar path could follow a symlink outside the agent workspace and return arbitrary file contents as a base64 data: URL in gateway responses.

ZDI-CAN-29311: OpenClaw Canvas Authentication Bypass Vulnerability – ABSTRACT ————————————- Trend Micro's Zero Day Initiative has identified a vulnerability affecting the following products: OpenClaw - OpenClaw – VULNERABILITY DETAILS ———————— Version tested: openclaw 2026.2.17 Platform tested: macOS 26.3

This vulnerability exists in Nuclio's Shell Runtime component, allowing attackers with function invocation permissions to inject malicious commands via HTTP request headers, execute arbitrary code with root privileges in function containers, steal ServiceAccount Tokens with cluster-admin level permissions, and ultimately achieve complete control over the entire Kubernetes cluster. Recommended CWE classification: CWE-78 (OS Command Injection). Nuclio Shell Runtime processes the X-Nuclio-Arguments HTTP header without validation or escaping, directly concatenating user …

A vulnerability in NLTK versions up to and including 3.9.2 allows arbitrary file read via path traversal in multiple CorpusReader classes, including WordListCorpusReader, TaggedCorpusReader, and BracketParseCorpusReader. These classes fail to properly sanitize or validate file paths, enabling attackers to traverse directories and access sensitive files on the server. This issue is particularly critical in scenarios where user-controlled file inputs are processed, such as in machine learning APIs, chatbots, or NLP …

The /api/server/shutdown endpoint allows termination of the Netmaker server process via syscall.SIGINT. This allows any user to repeatedly shut down the server, causing cyclic denial of service with approximately 3-second restart intervals.

An unsanitized qpack index can lead to an integer overflow, panicing in debug mode, accessing the wrong or no dynamic table entry in release mode. What does this mean for Firefox? Firefox runs Neqo in release mode. A malicious remote can cause its own QUIC connection to fail to use qpack, i.e. compression, or enter an inconsistent state. The remote can not crash Firefox, nor affect other QUIC connections.

The GET /1.0/certificates endpoint (non-recursive mode) returns URLs containing fingerprints for all certificates in the trust store, bypassing the per-object can_view authorization check that is correctly applied in the recursive path. Any authenticated identity — including restricted, non-admin users — can enumerate all certificate fingerprints, exposing the full set of trusted identities in the LXD deployment.

If exploited, this issue allows attackers to execute arbitrary JavaScript code in the Node.js process. It occurs when applications pass untrusted array callbacks to call_user_func_array(), a practice common in JSON-RPC setups and PHP-to-JavaScript porting layers. Since the library fails to properly sanitize inputs, this is considered a supplier defect rather than an integration error. This flaw has been exploited in practice, but it is not a "drive-by" vulnerability. It only …

The GET /api/v4/image/{filename} endpoint is vulnerable to unauthenticated SSRF through parameter injection in the file_type query parameter. An attacker can inject arbitrary query parameters into the internal request to pict-rs, including the proxy parameter which causes pict-rs to fetch arbitrary URLs.

GET /api/invoices/{id} only checks the role-based view_invoice permission but does not verify the requesting user has access to the invoice's customer. Any user with ROLE_TEAMLEAD (which grants view_invoice) can read all invoices in the system, including those belonging to customers assigned to other teams.

The UTF8DataInputJsonParser, which is used when parsing from a java.io.DataInput source, bypasses the maxNestingDepth constraint (default: 500) defined in StreamReadConstraints. A similar issue was found in ReaderBasedJsonParser. This allows a user to supply a JSON document with excessive nesting, which can cause a StackOverflowError when the structure is processed, leading to a Denial of Service (DoS). The related fix for com.fasterxml.jackson.core:jackson-core, CVE-2025-52999, was not fully applied to tools.jackson.core:jackson-core until the …

An attacker can manipulate the HTTP Host header on a password reset or account creation request. The confirmation link in the resulting email can then point to an attacker-controlled domain. Opening the link in the email is sufficient to pass the token to the attacker, who can then use it on the real IRRD instance to take over the account. A compromised account can then be used to modify RPSL …

What kind of vulnerability is it? Who is impacted? A Prototype Pollution is possible in immutable via the mergeDeep(), mergeDeepWith(), merge(), Map.toJS(), and Map.toObject() APIs.

When using streamSSE() in Streaming Helper, the event, id, and retry fields were not validated for carriage return (\r) or newline (\n) characters. Because the SSE protocol uses line breaks as field delimiters, this could allow injection of additional SSE fields within the same event frame if untrusted input was passed into these fields.

The setCookie() utility did not validate semicolons (;), carriage returns (\r), or newline characters (\n) in the domain and path options when constructing the Set-Cookie header. Because cookie attributes are delimited by semicolons, this could allow injection of additional cookie attributes if untrusted input was passed into these fields.

When using serveStatic together with route-based middleware protections (e.g. app.use('/admin/*', …)), inconsistent URL decoding allowed protected static resources to be accessed without authorization. The router used decodeURI, while serveStatic used decodeURIComponent. This mismatch allowed paths containing encoded slashes (%2F) to bypass middleware protections while still resolving to the intended filesystem path.

A broken access control vulnerability in the TUS protocol DELETE endpoint allows authenticated users with only Create permission to delete arbitrary files and directories within their scope, bypassing the intended Delete permission restriction. Any multi-user deployment where administrators explicitly restrict file deletion for certain users is affected.

fickling's UNSAFE_IMPORTS blocklist is missing at least 3 stdlib modules that provide direct arbitrary command execution: uuid, _osx_support, and _aix_support. These modules contain functions that internally call subprocess.Popen() or os.system() with attacker-controlled arguments. A malicious pickle file importing these modules passes both UnsafeImports and NonStandardImports checks.

fickling.always_check_safety() does not hook all pickle entry points. pickle.loads, _pickle.loads, and _pickle.load remain unprotected, enabling malicious payload execution despite global safety mode being enabled.

Dark Reader versions prior to 4.9.117 included a behavior where a website could request a style sheet from a locally running web server, for example http://localhost:8080/style.css, If an address was available and returned a text/css content type.

The actionSendActivationEmail() endpoint is accessible to unauthenticated users and does not require a permission check for pending users. An attacker with no prior access can trigger activation emails for any pending user account by knowing or guessing the user ID. If the attacker controls the target user’s email address, they can activate the account and gain access to the system. The vulnerability is not that anonymous access exists - there’s …

Concrete CMS below version 9.4.8 is vulnerable to Remote Code Execution by stored PHP object injection into the Express Entry List block via the columns parameter. An authenticated administrator can store attacker-controlled serialized data in block configuration fields that are later passed to unserialize() without class restrictions or integrity checks. The Concrete CMS security team thanks YJK ( @YJK0805 https://hackerone.com/yjk0805 ) of ZUSO ART https://zuso.ai/ for reporting.

Concrete CMS below version 9.4.8 is subject to CSRF by a Rogue Administrator using the Anti-Spam Allowlist Group Configuration via group_id parameter which can leads to a security bypass since changes are saved prior to checking the CSRF token. The Concrete CMS security team thanks z3rco for reporting

In Concrete CMS below version 9.4.8, a Cross-site Scripting (XSS) vulnerability exists in the "Legacy Form" block. An authenticated user with permissions to create or edit forms (e.g., a rogue administrator) can inject a persistent JavaScript payload into the options of a multiple-choice question (Checkbox List, Radio Buttons, or Select Box). This payload is then executed in the browser of any user who views the page containing the form. The …

In Concrete CMS below version 9.4.8, A stored Cross-site Scripting (XSS) vulnerability exists in the search block where page names and content are rendered without proper HTML encoding in search results. This allows authenticated, rogue administrators to inject malicious JavaScript through page names that executes when users search for and view those pages in search results. The Concrete CMS security team thanks zolpak for reporting.

In Concrete CMS below version 9.4.8, a user with permission to edit a page with element Legacy form can perform a stored XSS attack towards high-privilege accounts via the Question field. The Concrete CMS security team thanks minhnn42, namdi and quanlna2 from VCSLab-Viettel Cyber Security for reporting.

In Concrete CMS below version 9.4.8, a rogue administrator can add stored XSS via the Switch Language block. The Concrete CMS security team gave thanks M3dium for reporting.

A Cross-Site Scripting (XSS) vulnerability has been discovered in the General HTML Support feature. This vulnerability could be triggered by inserting specially crafted markup, leading to unauthorized JavaScript code execution, if the editor instance used an unsafe General HTML Support configuration. This vulnerability affects only installations where the editor configuration meets the following criteria:

A Cross-Site Scripting (XSS) vulnerability has been discovered in the General HTML Support feature. This vulnerability could be triggered by inserting specially crafted markup, leading to unauthorized JavaScript code execution, if the editor instance used an unsafe General HTML Support configuration. This vulnerability affects only installations where the editor configuration meets the following criteria:

The changedetection.io application allows users to specify XPath expressions as content filters via the include_filters field. These XPath expressions are processed using the elementpath library which implements XPath 3.0/3.1 specification. XPath 3.0 includes the unparsed-text() function which can read arbitrary files from the filesystem. The application does not validate or sanitize XPath expressions to block dangerous functions, allowing an attacker to read any file accessible to the application process.

A Zip Slip vulnerability in the backup restore functionality allows arbitrary file overwrite via path traversal in uploaded ZIP archives.

A reflected cross-site scripting (XSS) vulnerability was identified in the /rss/tag/ endpoint of changedetection.io. The tag_uuid path parameter is reflected directly in the HTTP response body without HTML escaping. Since Flask returns text/html by default for plain string responses, the browser parses and executes injected JavaScript. This vulnerability persists in version 0.54.1, which patched the related XSS in /rss/watch/ (CVE-2026-27645 / GHSA-mw8m-398g-h89w) but did not address the identical pattern in …

After upgrading the library from 1.5.2 to 1.6.0 (and the latest 1.6.5) it was noticed that previous tests involving passing a malicious JWT containing alg: none and an empty signature was passing the signature verification step without any changes to the application code when a failure was expected.

Missing Authentication for Critical Function (CWE-306) vulnerability in Apache Artemis, Apache ActiveMQ Artemis. An unauthenticated remote attacker can use the Core protocol to force a target broker to establish an outbound Core federation connection to an attacker-controlled rogue broker. This could potentially result in message injection into any queue and/or message exfiltration from any queue via the rogue broker. This impacts environments that allow both: Incoming Core protocol connections from …

Missing Authentication for Critical Function (CWE-306) vulnerability in Apache Artemis, Apache ActiveMQ Artemis. An unauthenticated remote attacker can use the Core protocol to force a target broker to establish an outbound Core federation connection to an attacker-controlled rogue broker. This could potentially result in message injection into any queue and/or message exfiltration from any queue via the rogue broker. This impacts environments that allow both: Incoming Core protocol connections from …

Apache ActiveMQ does not properly validate the remaining length field which may lead to an overflow during the decoding of malformed packets. When this integer overflow occurs, ActiveMQ may incorrectly compute the total Remaining Length and subsequently misinterpret the payload as multiple MQTT control packets which makes the broker susceptible to unexpected behavior when interacting with non-compliant clients. This behavior violates the MQTT v3.1.1 specification, which restricts Remaining Length to …

Apache ActiveMQ does not properly validate the remaining length field which may lead to an overflow during the decoding of malformed packets. When this integer overflow occurs, ActiveMQ may incorrectly compute the total Remaining Length and subsequently misinterpret the payload as multiple MQTT control packets which makes the broker susceptible to unexpected behavior when interacting with non-compliant clients. This behavior violates the MQTT v3.1.1 specification, which restricts Remaining Length to …

Apache ActiveMQ does not properly validate the remaining length field which may lead to an overflow during the decoding of malformed packets. When this integer overflow occurs, ActiveMQ may incorrectly compute the total Remaining Length and subsequently misinterpret the payload as multiple MQTT control packets which makes the broker susceptible to unexpected behavior when interacting with non-compliant clients. This behavior violates the MQTT v3.1.1 specification, which restricts Remaining Length to …

The time_calibrators crate attempted to exfiltrate .env files to a server that was in turn impersonating the legitimate timeapi.io service. The malicious crate had 1 version published on 2026-03-03 approximately 3 hours before removal and had no evidence of actual downloads. There were no crates depending on this crate on crates.io. Rust security response working group thanks cybergeek for finding and reporting this, and thanks to Emily Albini for co-ordinating …

It was reported time_calibrator contained malicious code, that would try to upload .env files to a server. The malicious crate had only 1 version published at 2026-02-28 and no evidence of actual usage. The crate was removed from crates.io and the user account was locked. There were no crates depending on this crate on crates.io. Rust security response working group thanks Gabriel Silva for finding and reporting this, and thanks …

When using @hono/node-server's static file serving together with route-based middleware protections (e.g. protecting /admin/*), inconsistent URL decoding can allow protected static resources to be accessed without authorization. In particular, paths containing encoded slashes (%2F) may be evaluated differently by routing/middleware matching versus static file path resolution, enabling a bypass where middleware does not run but the static file is still served.

An unauthenticated attacker can execute arbitrary OS commands on the server by injecting shell command substitution into the base64Url GET parameter. This can lead to full server compromise, data exfiltration (e.g., configuration secrets, internal keys, credentials), and service disruption.

A stored Cross-site Scripting (XSS) vulnerability exists on rendering TableBlock blocks within a StreamField. A user with access to create or edit pages containing TableBlock StreamField blocks is able to set specially-crafted class attributes on the block which run arbitrary JavaScript code when the page is viewed. When viewed by a user with higher privileges, this could lead to performing actions with that user's credentials. The vulnerability is not exploitable …

In simple words, some programs that use _.flatten or _.isEqual could be made to crash. Someone who wants to do harm may be able to do this on purpose. This can only be done if the program has special properties. It only works in Underscore versions up to 1.13.7. A more detailed explanation follows. In affected versions of Underscore, the _.flatten and _.isEqual functions use recursion without a depth limit. …

Sandbox media local-path validation accepted absolute paths under host tmp, even when those paths were outside the active sandbox root.

Sandbox media local-path validation accepted absolute paths under host tmp, even when those paths were outside the active sandbox root.

/api/query/sql allows users to run SQL directly, but it only checks basic auth, not admin rights, any logged-in user, even readers, can run any SQL query on the database.

This vulnerability only affects customers using Weave CNI (Container Network Interface) when configured through RKE templates. A flaw was discovered in Rancher versions from 2.5.0 up to and including 2.5.13 and from 2.6.0 up to and including 2.6.4, where a UI (user interface) issue with RKE templates does not include a value for the Weave password when Weave is chosen as the CNI. If a cluster is created based on …

The restricted pod security policy (PSP), provided in Rancher versions from 2.0 up to and including 2.6.3, has a deviation from the upstream restricted policy provided in Kubernetes, in which Rancher's PSP has runAsUser set to runAsAny, while upstream has runAsUser set to MustRunAsNonRoot. This allows containers to run as any user, including a privileged user (root), even when Rancher's restricted policy is enforced on a project or at cluster …

A bug has been identified in which permission changes in Azure AD are not reflected to users while they are logged in the Rancher UI. This would cause the users to retain their previous permissions in Rancher, even if they change groups on Azure AD, for example, to a lower privileged group, or are removed from a group, thus retaining their access to Rancher instead of losing it.

An issue was discovered in Rancher versions up to and including 2.5.15 and 2.6.6 where a flaw with authorization logic allows privilege escalation through cluster role template binding (CRTB) and project role template binding (PRTB). This issue does not affect the local cluster, it affects only downstream clusters. The vulnerability can be exploited by any user who has permissions to create/edit cluster role template bindings or project role template bindings …

It was discovered that in Rancher versions up to and including 2.5.12 and 2.6.3 there is a failure to properly sanitize credentials in cluster template answers. This failure can lead to plaintext storage and exposure of credentials, passwords and API tokens. The exposed credentials are visible in Rancher to authenticated Cluster Owners, Cluster Members, Project Owners and Project Members on the endpoints /v1/management.cattle.io.clusters, /v3/clusters and /k8s/clusters/local/apis/management.cattle.io/v3/clusters. Sensitive fields are now …

A vulnerability was discovered in Rancher 2.2.0 through the aforementioned patched versions, where cloud credentials weren't being properly validated through the Rancher API. Specifically through a proxy designed to communicate with cloud providers. Any Rancher user that was logged-in and aware of a cloud-credential ID that was valid for a given cloud provider, could call that cloud provider's API through the proxy API, and the cloud-credential would be attached. The …

A vulnerability has been identified within the Rancher Backup Operator, resulting in the leakage of S3 tokens (both accessKey and secretKey) into the rancher-backup-operator pod's logs. Specifically, the S3 accessKey and secretKey are exposed in the pod's logs under the following logging level conditions: | Variable Exposed | Logging Level Condition | —————— | ————————- | | accessKey | trace: false (default), and debug: false (default) | | secretKey | …

picklescan v1.0.3 blocks profile.Profile.run and profile.Profile.runctx but does NOT block the module-level profile.run() function. A malicious pickle calling profile.run(statement) achieves arbitrary code execution via exec() while picklescan reports 0 issues. This is because the blocklist entry "Profile.run" does not match the pickle global name "run".

pkgutil.resolve_name() is a Python stdlib function that resolves any "module:attribute" string to the corresponding Python object at runtime. By using pkgutil.resolve_name as the first REDUCE call in a pickle, an attacker can obtain a reference to ANY blocked function (e.g., os.system, builtins.exec, subprocess.call) without that function appearing in the pickle's opcodes. picklescan only sees pkgutil.resolve_name (which is not blocked) and misses the actual dangerous function entirely. This defeats picklescan's entire …

picklescan v1.0.3 (latest) does not block at least 7 Python standard library modules that provide direct arbitrary command execution or code evaluation. A malicious pickle file importing these modules is reported as having 0 issues (CLEAN scan). This enables remote code execution that bypasses picklescan entirely.

OpenViking versions 0.2.1 and prior, fixed in commit 46b3e76, contain a path traversal vulnerability in the .ovpack import handling that allows attackers to write files outside the intended import directory. Attackers can craft malicious ZIP archives with traversal sequences, absolute paths, or drive prefixes in member names to overwrite or create arbitrary files with the importing process privileges.

Multiple Reflected Cross-Site Scripting (XSS) vulnerabilities in OpenSTAManager v2.9.8 allow unauthenticated attackers to execute arbitrary JavaScript code in the context of other users' browsers through crafted URL parameters, potentially leading to session hijacking, credential theft, and unauthorized actions. Vulnerable Parameter: righe (GET)

A privilege escalation and authentication bypass vulnerability in OpenSTAManager allows any attacker to arbitrarily change a user's group (idgruppo) by directly calling modules/utenti/actions.php. This can promote an existing account (e.g. agent) into the Amministratori group as well as demote any user including existing administrators.

A missing group-sender authorization check in the Zalo plugin allowed unauthorized GROUP messages to enter agent dispatch paths in configurations intended to restrict group traffic.

openclaw web tools strict URL fetch paths could lose DNS pinning when environment proxy variables are configured (HTTP_PROXY/HTTPS_PROXY/ALL_PROXY, including lowercase variants). In affected builds, strict URL checks (for example web_fetch and citation redirect resolution) validated one destination during SSRF guard checks, but runtime connection routing could proceed through an env-proxy dispatcher.

openclaw web tools strict URL fetch paths could lose DNS pinning when environment proxy variables are configured (HTTP_PROXY/HTTPS_PROXY/ALL_PROXY, including lowercase variants). In affected builds, strict URL checks (for example web_fetch and citation redirect resolution) validated one destination during SSRF guard checks, but runtime connection routing could proceed through an env-proxy dispatcher.

Twilio webhook replay events could bypass voice-call manager dedupe because normalized event IDs were randomized per parse. A replayed event could be treated as new and trigger duplicate or stale call-state transitions.

Twilio webhook replay events could bypass voice-call manager dedupe because normalized event IDs were randomized per parse. A replayed event could be treated as new and trigger duplicate or stale call-state transitions.

The voice-call Twilio webhook path accepted replay/dedupe identity from unsigned request metadata (i-twilio-idempotency-token), enabling replayed signed requests to bypass replay detection and manager dedupe by mutating only that header.

channels..groups..toolsBySender could match a privileged sender policy using a colliding mutable identity value (for example senderName or senderUsername) when deployments used untyped keys. The fix introduces explicit typed sender keys (id:, e164:, username:, name:), keeps legacy untyped keys on a deprecated ID-only path, and adds regression coverage to prevent cross-identifier collisions.

channels..groups..toolsBySender could match a privileged sender policy using a colliding mutable identity value (for example senderName or senderUsername) when deployments used untyped keys. The fix introduces explicit typed sender keys (id:, e164:, username:, name:), keeps legacy untyped keys on a deprecated ID-only path, and adds regression coverage to prevent cross-identifier collisions.

In openclaw allowlist mode, tools.exec.safeBins trusted PATH-derived directories for safe-bin resolution. A same-name binary placed in a trusted PATH directory could satisfy safe-bin checks and execute.

In OpenClaw, tools.exec.safeBins validation for sort could be bypassed via GNU long-option abbreviations in allowlist mode, allowing approval-free execution paths that should require approval.

In OpenClaw, tools.exec.safeBins validation for sort could be bypassed via GNU long-option abbreviations in allowlist mode, allowing approval-free execution paths that should require approval.

When tools.exec.safeBins contained a binary without an explicit safe-bin profile, OpenClaw used a permissive generic fallback profile. In allowlist mode, that could let interpreter-style binaries (for example python3, node, ruby) execute inline payloads via flags like -c. This requires explicit operator configuration to add such binaries to safeBins, so impact is limited to non-default/misconfigured deployments.

A missing sender-authorization check in Telegram message_reaction handling allowed unauthorized users to trigger reaction-derived system events.

In openclaw up to and including 2026.2.23 (latest npm release as of February 25, 2026), system.run shell-wrapper inputs could present misleading approval/display text while still carrying hidden positional argv payloads that execute at runtime.

In openclaw up to and including 2026.2.23 (latest npm release as of February 25, 2026), system.run shell-wrapper inputs could present misleading approval/display text while still carrying hidden positional argv payloads that execute at runtime.

In openclaw@2026.2.24, approval-bound system.run on node hosts could be influenced by mutable symlink cwd targets between approval and execution.

In openclaw@2026.2.24, approval-bound system.run on node hosts could be influenced by mutable symlink cwd targets between approval and execution.

In OpenClaw system.run allowlist mode, shell-wrapper analysis could be bypassed by splitting command substitution as $\ + newline + ( inside double quotes. Analysis treated the payload as allowlisted (for example /bin/echo), while shell runtime folded the line continuation into $(…) and executed non-allowlisted subcommands.

In OpenClaw system.run allowlist mode, shell-wrapper analysis could be bypassed by splitting command substitution as $\ + newline + ( inside double quotes. Analysis treated the payload as allowlisted (for example /bin/echo), while shell runtime folded the line continuation into $(…) and executed non-allowlisted subcommands.

In openclaw versions 2026.2.22 and 2026.2.23, the optional synology-chat channel plugin had an authorization fail-open condition: when dmPolicy was allowlist and allowedUserIds was empty/unset, unauthorized senders were still allowed through to agent dispatch. This is assessed as medium severity because it requires channel/plugin setup and Synology sender access, but can still trigger downstream agent/tool actions.

In openclaw versions 2026.2.22 and 2026.2.23, the optional synology-chat channel plugin had an authorization fail-open condition: when dmPolicy was allowlist and allowedUserIds was empty/unset, unauthorized senders were still allowed through to agent dispatch. This is assessed as medium severity because it requires channel/plugin setup and Synology sender access, but can still trigger downstream agent/tool actions.

OpenClaw Slack monitor handled reaction_* and pin_* non-message events before applying sender-policy checks consistently. In affected versions, these events could be added to system-event context even when sender policy would not normally allow them.

OpenClaw Slack monitor handled reaction_* and pin_* non-message events before applying sender-policy checks consistently. In affected versions, these events could be added to system-event context even when sender policy would not normally allow them.

In a narrow Signal reaction-notification path, reaction-only inbound events could enqueue a status event before sender access checks were applied.

In a narrow Signal reaction-notification path, reaction-only inbound events could enqueue a status event before sender access checks were applied.

system.run environment sanitization allowed shell-startup env overrides (HOME, ZDOTDIR) that can execute attacker-controlled startup files before allowlist-evaluated command bodies.

system.run environment sanitization allowed shell-startup env overrides (HOME, ZDOTDIR) that can execute attacker-controlled startup files before allowlist-evaluated command bodies.

The shell environment fallback path could invoke an attacker-controlled shell when SHELL was inherited from an untrusted host environment. In affected builds, shell-env loading used $SHELL -l -c 'env -0' without validating that SHELL points to a trusted executable. In threat-model terms, this requires local environment compromise or untrusted startup environment injection first; it is not a remote pre-auth path. The hardening patch validates SHELL as an absolute normalized executable, …

The shell environment fallback path could invoke an attacker-controlled shell when SHELL was inherited from an untrusted host environment. In affected builds, shell-env loading used $SHELL -l -c 'env -0' without validating that SHELL points to a trusted executable. In threat-model terms, this requires local environment compromise or untrusted startup environment injection first; it is not a remote pre-auth path. The hardening patch validates SHELL as an absolute normalized executable, …

Concurrent updateRegistry/removeRegistryEntry operations for sandbox containers and browsers could lose updates or resurrect removed entries under race conditions. The registry writes were read-modify-write in a window with no locking and permissive fallback parsing, so concurrent registry updates could produce stale snapshots and overwrite each other. That desyncs sandbox state and can affect sandbox list, sandbox prune, and sandbox recreate –all behavior.

Concurrent updateRegistry/removeRegistryEntry operations for sandbox containers and browsers could lose updates or resurrect removed entries under race conditions. The registry writes were read-modify-write in a window with no locking and permissive fallback parsing, so concurrent registry updates could produce stale snapshots and overwrite each other. That desyncs sandbox state and can affect sandbox list, sandbox prune, and sandbox recreate –all behavior.

In openclaw up to and including 2026.2.23 (latest npm release as of February 24, 2026), sandbox bind-source validation could be bypassed when a bind source used a symlinked parent plus a non-existent leaf path.

In openclaw up to and including 2026.2.23 (latest npm release as of February 24, 2026), sandbox bind-source validation could be bypassed when a bind source used a symlinked parent plus a non-existent leaf path.

OpenClaw accepted prototype-reserved keys in runtime /debug set override object values (proto, constructor, prototype).

OpenClaw accepted prototype-reserved keys in runtime /debug set override object values (proto, constructor, prototype).

In authenticated non-owner DM sessions, a narrow tool-invocation path could reach broader-than-intended owner-only gateway actions.

When sort is explicitly added to tools.exec.safeBins (non-default), the –compress-program option can invoke an external helper and bypass the intended safe-bin approval constraints in allowlist mode.

When sort is explicitly added to tools.exec.safeBins (non-default), the –compress-program option can invoke an external helper and bypass the intended safe-bin approval constraints in allowlist mode.

In openclaw versions up to and including 2026.2.22-2, a non-default exec-approval configuration could allow a skill-name collision to bypass an ask=on-miss prompt. When autoAllowSkills=true, a path-scoped executable such as ./skill-bin could resolve to basename skill-bin, satisfy the skills allowlist segment, and run without prompting for approval.

In openclaw@2026.3.1, node system.run approval-path hardening rewrote wrapper command argv in a way that changed execution semantics. A command shown/approved as a shell payload (for example echo SAFE) could execute a different local script when wrapper argv were rewritten.

In openclaw@2026.3.1, node system.run approval-path hardening rewrote wrapper command argv in a way that changed execution semantics. A command shown/approved as a shell payload (for example echo SAFE) could execute a different local script when wrapper argv were rewritten.

A client authenticated with a shared gateway token could connect as role=node without device identity/pairing, then call node.event to trigger agent.request and voice.transcript flows.

A client authenticated with a shared gateway token could connect as role=node without device identity/pairing, then call node.event to trigger agent.request and voice.transcript flows.

When Nextcloud Talk webhook signing was valid, replayed requests could be accepted without durable replay suppression, allowing duplicate inbound processing after replay-window expiry or process restart.

When Nextcloud Talk webhook signing was valid, replayed requests could be accepted without durable replay suppression, allowing duplicate inbound processing after replay-window expiry or process restart.

In OpenClaw MSTeams media download flows, redirect handling could bypass configured mediaAllowHosts checks in specific attachment paths. Redirect chains were not consistently constrained to allowlisted targets before accepting fetched content.

In OpenClaw MSTeams media download flows, redirect handling could bypass configured mediaAllowHosts checks in specific attachment paths. Redirect chains were not consistently constrained to allowlisted targets before accepting fetched content.

OpenClaw hook mapping transforms could be loaded via absolute paths or .. traversal, allowing arbitrary JavaScript module loading/execution in the gateway process when an attacker can modify hooks configuration.

OpenClaw hook mapping transforms could be loaded via absolute paths or .. traversal, allowing arbitrary JavaScript module loading/execution in the gateway process when an attacker can modify hooks configuration.

When tokenless Tailscale auth is enabled, OpenClaw should only allow forwarded-header auth for Control UI websocket authentication on trusted hosts. In affected versions, that tokenless path could also be used by HTTP gateway auth call sites, which could bypass token/password requirements for HTTP routes in trusted-network deployments.

When tokenless Tailscale auth is enabled, OpenClaw should only allow forwarded-header auth for Control UI websocket authentication on trusted hosts. In affected versions, that tokenless path could also be used by HTTP gateway auth call sites, which could bypass token/password requirements for HTTP routes in trusted-network deployments.

system.run exec allowlist analysis treated wrapper binaries as the effective executable and did not fully unwrap env/shell-dispatch wrappers. This allowed wrapper-smuggled payloads (for example env bash -lc …) to satisfy an allowlist entry for the wrapper while executing non-allowlisted commands.

system.run exec allowlist analysis treated wrapper binaries as the effective executable and did not fully unwrap env/shell-dispatch wrappers. This allowed wrapper-smuggled payloads (for example env bash -lc …) to satisfy an allowlist entry for the wrapper while executing non-allowlisted commands.

A wrapper-depth parsing mismatch in system.run allowed nested transparent dispatch wrappers (for example repeated /usr/bin/env) to suppress shell-wrapper detection while still matching allowlist resolution. In security=allowlist + ask=on-miss, this could bypass the expected approval prompt for shell execution.

A wrapper-depth parsing mismatch in system.run allowed nested transparent dispatch wrappers (for example repeated /usr/bin/env) to suppress shell-wrapper detection while still matching allowlist resolution. In security=allowlist + ask=on-miss, this could bypass the expected approval prompt for shell execution.

OpenClaw allowed dangerous process-control environment variables from env.vars (for example NODE_OPTIONS, LD_, DYLD_) to flow into gateway service runtime environments, enabling startup-time code execution in the OpenClaw process context.

OpenClaw allowed dangerous process-control environment variables from env.vars (for example NODE_OPTIONS, LD_, DYLD_) to flow into gateway service runtime environments, enabling startup-time code execution in the OpenClaw process context.

commands.allowFrom is documented as a sender authorization allowlist for commands/directives, but command authorization could include ctx.From (conversation identity) as a sender candidate. When commands.allowFrom contained conversation-like identifiers (for example Discord channel:<id> or WhatsApp group JIDs), command/directive authorization could be granted to participants in that conversation instead of only the intended sender identity.

This issue is a browser-origin WebSocket auth chain on local loopback deployments using password auth. It is serious, but conditional: an attacker must get the user to open a malicious page and then successfully guess the gateway password.

This issue is a browser-origin WebSocket auth chain on local loopback deployments using password auth. It is serious, but conditional: an attacker must get the user to open a malicious page and then successfully guess the gateway password.

OpenClaw avatar handling allowed a symlink traversal path that could expose local files outside an agent workspace through gateway avatar surfaces.

OpenClaw avatar handling allowed a symlink traversal path that could expose local files outside an agent workspace through gateway avatar surfaces.

The sandbox browser entrypoint launched x11vnc without authentication (-nopw) for noVNC observer sessions. OpenClaw-managed runtime flow publishes the noVNC port to host loopback only (127.0.0.1), so default exposure is local to the host unless operators explicitly expose the port more broadly (or run the image standalone with broad port publishing).

The sandbox browser entrypoint launched x11vnc without authentication (-nopw) for noVNC observer sessions. OpenClaw-managed runtime flow publishes the noVNC port to host loopback only (127.0.0.1), so default exposure is local to the host unless operators explicitly expose the port more broadly (or run the image standalone with broad port publishing).

tools.exec.safeBins allowlist checks could be bypassed by PATH-hijacked binaries, allowing execution of attacker-controlled trojan binaries under an allowlisted executable name.

tools.exec.safeBins allowlist checks could be bypassed by PATH-hijacked binaries, allowing execution of attacker-controlled trojan binaries under an allowlisted executable name.

A path confinement bypass in OpenClaw ZIP extraction allowed writes outside the intended destination when a pre-existing symlink was present under the extraction root.

ZIP extraction in OpenClaw could be raced into writing outside the intended destination directory via parent-directory symlink rebind between validation and write.

ZIP extraction in OpenClaw could be raced into writing outside the intended destination directory via parent-directory symlink rebind between validation and write.

A path-confinement bypass in browser output handling allowed writes outside intended roots in openclaw versions up to and including 2026.3.1. The fix unifies root-bound, file-descriptor-verified write semantics and canonical path-boundary validation across browser output and related install/skills write paths.

A path-confinement bypass in browser output handling allowed writes outside intended roots in openclaw versions up to and including 2026.3.1. The fix unifies root-bound, file-descriptor-verified write semantics and canonical path-boundary validation across browser output and related install/skills write paths.

In Telegram DM mode, inbound media was downloaded and written to disk before sender authorization checks completed. An unauthorized sender could trigger inbound media download/write activity (including media groups) even when DM access should be denied.

stageSandboxMedia allowed destination symlink traversal during media staging, which could overwrite files outside the sandbox workspace root.

stageSandboxMedia allowed destination symlink traversal during media staging, which could overwrite files outside the sandbox workspace root.

shell-env fallback trusted prefix-based executable paths for $SHELL, allowing execution of attacker-controlled binaries in local/runtime-env influence scenarios.

shell-env fallback trusted prefix-based executable paths for $SHELL, allowing execution of attacker-controlled binaries in local/runtime-env influence scenarios.

Sandboxed sessions_spawn(runtime="acp") could bypass sandbox inheritance and initialize host-side ACP runtime. The fix now fail-closes ACP spawn from sandboxed requester sessions and rejects sandbox="require" for runtime="acp".

A sandbox path validation bypass in openclaw allows host file reads outside sandboxRoot via the media path fallback tmp flow when the fallback tmp root is a symlink alias.

In openclaw<=2026.2.23, safe-bin trust in allowlist mode relied on static default trusted directories that included package-manager paths (notably /opt/homebrew/bin and /usr/local/bin). When a same-name binary (for example jq) is placed in one of those trusted default directories, safe-bin evaluation can be satisfied and execute the attacker-controlled binary.

In openclaw<=2026.2.23, safe-bin trust in allowlist mode relied on static default trusted directories that included package-manager paths (notably /opt/homebrew/bin and /usr/local/bin). When a same-name binary (for example jq) is placed in one of those trusted default directories, safe-bin evaluation can be satisfied and execute the attacker-controlled binary.

A paired node device could reconnect with spoofed platform/deviceFamily metadata and broaden node command policy eligibility because reconnect metadata was accepted from the client while these fields were not bound into the device-auth signature.

A paired node device could reconnect with spoofed platform/deviceFamily metadata and broaden node command policy eligibility because reconnect metadata was accepted from the client while these fields were not bound into the device-auth signature.

OpenClaw accepted camera.snap / camera.clip node payload url fields and downloaded them on the gateway/agent host without binding downloads to the resolved node host. In OpenClaw's documented trust model, paired nodes are in the same operator trust boundary, so this is scoped as medium-severity hardening. A malicious or compromised paired node could still steer gateway-host fetches during camera URL retrieval.

In sandboxed runs, native prompt image auto-load did not honor tools.fs.workspaceOnly=true. This optional hardening setting is not enabled by default. When operators enabled it, prompt text could still reference mounted out-of-workspace image paths (for example /agent/secret.png) and load those image bytes for vision-capable model input.

In openclaw MS Teams file-consent flow, pending uploads were authorized by uploadId alone. fileConsent/invoke did not verify the invoke conversation against the conversation that created the pending upload.

Microsoft Teams media handling used mixed fetch paths for Graph metadata/content and attachment auth-retry flows. Some paths bypassed the shared SSRF guard model and created inconsistent host/DNS enforcement across redirect/fetch hops.

On macOS node-host, optional exec-approval allowlist mode previously treated basename-only entries (for example echo) as trusted command matches. This could allow a same-name local binary (for example ./echo) to run without approval under security=allowlist + ask=on-miss.

On macOS node-host, optional exec-approval allowlist mode previously treated basename-only entries (for example echo) as trusted command matches. This could allow a same-name local binary (for example ./echo) to run without approval under security=allowlist + ask=on-miss.

The affected surface is the OpenClaw macOS app onboarding flow, and the macOS app is currently in beta. In that beta onboarding flow, Anthropic OAuth used the PKCE code_verifier value as OAuth state, exposing that secret in front-channel URL state.

In some opt-in sandbox configurations, the experimental apply_patch tool did not consistently apply workspace-only checks to mounted paths (for example /agent/…).

In some opt-in sandbox configurations, the experimental apply_patch tool did not consistently apply workspace-only checks to mounted paths (for example /agent/…).

In openclaw@2026.3.1, the Discord voice transcript path called agentCommand(…) without senderIsOwner, and agentCommand defaults missing senderIsOwner to true. This could allow a non-owner voice participant in the same channel to reach owner-only tool surfaces (gateway, cron) during voice transcript turns.

In openclaw@2026.3.1, the Discord voice transcript path called agentCommand(…) without senderIsOwner, and agentCommand defaults missing senderIsOwner to true. This could allow a non-owner voice participant in the same channel to reach owner-only tool surfaces (gateway, cron) during voice transcript turns.

In OpenClaw <= 2026.2.24, Discord direct-message reaction notifications did not consistently apply the same DM authorization checks (dmPolicy / allowFrom) that are enforced for normal DM message ingress. In restrictive DM setups, a non-allowlisted Discord user who can react to a bot-authored DM message could still enqueue a reaction-derived system event in the session. This is a reaction-only ingress inconsistency. By itself it does not directly execute commands; practical impact …

In OpenClaw <= 2026.2.24, Discord direct-message reaction notifications did not consistently apply the same DM authorization checks (dmPolicy / allowFrom) that are enforced for normal DM message ingress. In restrictive DM setups, a non-allowlisted Discord user who can react to a bot-authored DM message could still enqueue a reaction-derived system event in the session. This is a reaction-only ingress inconsistency. By itself it does not directly execute commands; practical impact …

Sandbox browser container launched Chromium with –no-sandbox by default, disabling Chromium's OS-level sandbox protections.

Sandbox browser container launched Chromium with –no-sandbox by default, disabling Chromium's OS-level sandbox protections.

BlueBubbles webhook auth in the optional beta iMessage plugin allowed a passwordless fallback path. In some reverse-proxy/local routing setups, this could allow unauthenticated webhook events.

BlueBubbles webhook auth in the optional beta iMessage plugin allowed a passwordless fallback path. In some reverse-proxy/local routing setups, this could allow unauthenticated webhook events.

OpenClaw Windows Scheduled Task script generation allowed unsafe argument handling in generated gateway.cmd files. In vulnerable versions, cmd metacharacter-only values could be emitted without safe quoting/escaping, which could lead to unintended command execution when the scheduled task runs.

OpenClaw Windows Scheduled Task script generation allowed unsafe argument handling in generated gateway.cmd files. In vulnerable versions, cmd metacharacter-only values could be emitted without safe quoting/escaping, which could lead to unintended command execution when the scheduled task runs.

When iMessage remote attachment fetching is enabled (channels.imessage.remoteHost), stageSandboxMedia accepted arbitrary absolute paths and used SCP to copy them into local staging. If a non-attachment path reaches this flow, files outside expected iMessage attachment directories on the remote host can be staged.

When iMessage remote attachment fetching is enabled (channels.imessage.remoteHost), stageSandboxMedia accepted arbitrary absolute paths and used SCP to copy them into local staging. If a non-attachment path reaches this flow, files outside expected iMessage attachment directories on the remote host can be staged.

OpenClaw’s Feishu media download flow used untrusted Feishu media keys (imageKey / fileKey) when building temporary file paths in extensions/feishu/src/media.ts. Because those keys were interpolated directly into temp-file paths, traversal segments could escape the temp directory and redirect writes outside os.tmpdir().

OpenClaw’s Feishu media download flow used untrusted Feishu media keys (imageKey / fileKey) when building temporary file paths in extensions/feishu/src/media.ts. Because those keys were interpolated directly into temp-file paths, traversal segments could escape the temp directory and redirect writes outside os.tmpdir().

The HTML session exporter (src/auto-reply/reply/export-html/template.js) interpolates img.mimeType directly into <img src="data:…"> attributes without validation or escaping. A crafted mimeType value (e.g., x" onerror="alert(1)) can break out of the attribute context and execute arbitrary JavaScript.

The HTML session exporter (src/auto-reply/reply/export-html/template.js) interpolates img.mimeType directly into <img src="data:…"> attributes without validation or escaping. A crafted mimeType value (e.g., x" onerror="alert(1)) can break out of the attribute context and execute arbitrary JavaScript.

A successful exploit can expose local secrets and credentials readable by the OpenClaw process user, including API keys and private config material.

A successful exploit can expose local secrets and credentials readable by the OpenClaw process user, including API keys and private config material.

A client using shared gateway auth could attach an unpaired device identity and request elevated operator scopes (including operator.admin) before pairing approval, enabling privilege escalation.

The tar.bz2 installer path in src/agents/skills-install-download.ts used shell tar preflight/extract logic that did not share the same hardening guarantees as the centralized archive extractor. This allowed crafted .tar.bz2 archives to bypass special-entry blocking and extracted-size guardrails enforced on other archive paths, causing local availability impact during skill install.

OpenClaw shell-env fallback trusted startup environment values and could execute attacker-influenced login-shell startup paths before loading env keys.

OpenClaw tools.exec.safeBins had a stdin-only policy bypass for grep. If pattern input was supplied through -e / –regexp, the validator consumed the pattern as a flag value and still allowed one positional operand. That positional could be a bare filename like .env.

OpenClaw tools.exec.safeBins had a stdin-only policy bypass for grep. If pattern input was supplied through -e / –regexp, the validator consumed the pattern as a flag value and still allowed one positional operand. That positional could be a bare filename like .env.

Auth-secret dual-use across security domains (gateway auth and prompt metadata hashing). Hash outputs are visible to third-party model providers in system prompts. No direct plaintext token disclosure. Practical risk is highest when operators use weak gateway tokens and leave owner hash secret unset.

Auth-secret dual-use across security domains (gateway auth and prompt metadata hashing). Hash outputs are visible to third-party model providers in system prompts. No direct plaintext token disclosure. Practical risk is highest when operators use weak gateway tokens and leave owner hash secret unset.

OpenClaw plugins/extensions run in-process and are treated as trusted code. This advisory tracks trust-boundary clarification around plugin runtime command execution (runtime.system.runCommandWithTimeout).

A local process can capture the OpenClaw Gateway auth token from Chrome CDP probe traffic on loopback.

A local process can capture the OpenClaw Gateway auth token from Chrome CDP probe traffic on loopback.

OpenClaw used left-most X-Forwarded-For values when requests came from configured trusted proxies. In proxy chains that append/preserve header values, this could let attacker-controlled header content influence security decisions tied to client IP.

OpenClaw used left-most X-Forwarded-For values when requests came from configured trusted proxies. In proxy chains that append/preserve header values, this could let attacker-controlled header content influence security decisions tied to client IP.

A command injection vulnerability exists in OpenClaw’s Linux systemd unit generation path. When rendering Environment= entries, attacker-controlled values are not rejected for CR/LF, and systemdEscapeArg() uses an incorrect whitespace-matching regex. This allows newline injection to break out of an Environment= line and inject standalone systemd directives (for example, ExecStartPre=). On service restart, the injected command is executed, resulting in local arbitrary command execution (local RCE) under the gateway service user.

A command injection vulnerability exists in OpenClaw’s Linux systemd unit generation path. When rendering Environment= entries, attacker-controlled values are not rejected for CR/LF, and systemdEscapeArg() uses an incorrect whitespace-matching regex. This allows newline injection to break out of an Environment= line and inject standalone systemd directives (for example, ExecStartPre=). On service restart, the injected command is executed, resulting in local arbitrary command execution (local RCE) under the gateway service user.

When an attacker can cause a transform module path to reference a symlinked entry that resolves outside the trusted transform directory, the gateway may import and execute unintended JavaScript with gateway-process privileges.

On Windows, the Lobster extension previously retried certain spawn failures (ENOENT/EINVAL) with shell: true for wrapper compatibility. In that fallback path, tool-provided arguments could be interpreted by cmd.exe if fallback was triggered.

On Windows, the Lobster extension previously retried certain spawn failures (ENOENT/EINVAL) with shell: true for wrapper compatibility. In that fallback path, tool-provided arguments could be interpreted by cmd.exe if fallback was triggered.

system.run allowed SHELLOPTS + PS4 environment injection to trigger command substitution during bash -lc xtrace expansion before the allowlisted command body executed.

system.run allowed SHELLOPTS + PS4 environment injection to trigger command substitution during bash -lc xtrace expansion before the allowlisted command body executed.

The exported session HTML viewer allowed stored XSS when untrusted session content included raw HTML markdown tokens or unescaped metadata fields.

OpenClaw webhook handlers for BlueBubbles and Google Chat accepted and parsed request bodies before authentication and signature checks on vulnerable releases. This allowed unauthenticated clients to hold parser work open with slow/oversized request bodies and degrade availability (slow-request DoS).

OpenClaw webhook handlers for BlueBubbles and Google Chat accepted and parsed request bodies before authentication and signature checks on vulnerable releases. This allowed unauthenticated clients to hold parser work open with slow/oversized request bodies and degrade availability (slow-request DoS).

Three Dockerfiles in scripts/docker/ and scripts/e2e/ lack a USER directive, meaning all processes run as uid 0 (root). If any process is compromised, the attacker has root inside the container, making container breakout significantly easier. Partial fix (2026-02-08): Commit 28e1a65e added USER sandbox to Dockerfile.sandbox and Dockerfile.sandbox-browser. The E2E/test Dockerfiles listed below remain unpatched. Affected components: scripts/e2e/Dockerfile scripts/e2e/Dockerfile.qr-import scripts/docker/install-sh-e2e/Dockerfile scripts/docker/install-sh-nonroot/Dockerfile (runs as app but with NOPASSWD sudo — see related …

In OpenClaw's macOS node-host path, system.run allowlist parsing in security=allowlist mode failed to reject command substitution tokens when they appeared inside double-quoted shell text. Because of that gap, payloads like echo "ok $(id)" could be treated as allowlist hits (first executable token echo) while still executing non-allowlisted subcommands through shell substitution.

In OpenClaw's macOS node-host path, system.run allowlist parsing in security=allowlist mode failed to reject command substitution tokens when they appeared inside double-quoted shell text. Because of that gap, payloads like echo "ok $(id)" could be treated as allowlist hits (first executable token echo) while still executing non-allowlisted subcommands through shell substitution.

An attacker could split failed hook-auth attempts across both address forms and effectively double the brute-force budget from 20 to 40 attempts per 60-second window.

Gateway plugin route auth protection for /api/channels could be bypassed using encoded dot-segment traversal (for example ..%2f) in path variants that plugin handlers normalize.

Gateway plugin route auth protection for /api/channels could be bypassed using encoded dot-segment traversal (for example ..%2f) in path variants that plugin handlers normalize.

tools.exec allowlist/safe-bins evaluation could diverge from runtime execution for wrapper commands using GNU env -S/–split-string semantics. This allowed policy checks to treat a command as a benign safe-bin invocation while runtime executed a different payload.

Encoded alternate-path requests could bypass plugin route auth checks for /api/channels/* due to canonicalization depth mismatch in vulnerable builds.

Encoded alternate-path requests could bypass plugin route auth checks for /api/channels/* due to canonicalization depth mismatch in vulnerable builds.

The Lobster extension tool execution path used a Windows shell fallback (shell: true) after spawn failures (EINVAL/ENOENT). In that fallback path, shell metacharacters in command arguments can be interpreted by the shell, enabling command injection.

The Lobster extension tool execution path used a Windows shell fallback (shell: true) after spawn failures (EINVAL/ENOENT). In that fallback path, shell metacharacters in command arguments can be interpreted by the shell, enabling command injection.

This advisory tracks a defense-in-depth hardening for canvas routes. In mixed-trust or network-visible deployments, prior canvas auth/fallback behavior could broaden access beyond intended boundaries.

When the optional Chrome extension relay is enabled, /extension accepted unauthenticated WebSocket upgrades while /json/* and /cdp required auth.

In affected releases, when an operator explicitly enabled gateway.controlUi.allowInsecureAuth: true and exposed the gateway over plaintext HTTP, Control UI authentication could permit privileged operator access without the intended device identity + pairing guarantees. This required an insecure deployment choice and credential exposure risk (for example, plaintext transit or prior token leak). It was fixed on main in commit 40a292619e1f2be3a3b1db663d7494c9c2dc0abf (PR #20684).

In affected releases, when an operator explicitly enabled gateway.controlUi.allowInsecureAuth: true and exposed the gateway over plaintext HTTP, Control UI authentication could permit privileged operator access without the intended device identity + pairing guarantees. This required an insecure deployment choice and credential exposure risk (for example, plaintext transit or prior token leak). It was fixed on main in commit 40a292619e1f2be3a3b1db663d7494c9c2dc0abf (PR #20684).

In allowlist mode, system.run guardrails could be bypassed through env -S, causing policy-analysis/runtime-execution mismatch for shell wrapper payloads.

In allowlist mode, system.run guardrails could be bypassed through env -S, causing policy-analysis/runtime-execution mismatch for shell wrapper payloads.

A workspace-only file-system guard mismatch allowed @-prefixed absolute paths to bypass boundary validation in some tool path checks.

A workspace-only file-system guard mismatch allowed @-prefixed absolute paths to bypass boundary validation in some tool path checks.

A trusted-proxy Control UI pairing bypass accepted client.id=control-ui without device identity checks. The bypass did not require operator role, so an authenticated node role session could connect unpaired and reach node event methods.

A trusted-proxy Control UI pairing bypass accepted client.id=control-ui without device identity checks. The bypass did not require operator role, so an authenticated node role session could connect unpaired and reach node event methods.

OpenClaw's SSRF IP classifier did not treat IPv6 multicast literals (ff00::/8) as blocked/private-internal. This allowed literal multicast hosts to pass SSRF preflight checks.

Feishu allowlist authorization could be bypassed by display-name collision.

Feishu allowlist authorization could be bypassed by display-name collision.

OpenClaw supports Discord allowlists using either user IDs or names/tags. Name/tag matching depends on slug normalization, so different user tags can collide to the same slug and unintentionally satisfy a name-based allowlist entry.

A command injection vulnerability existed in Windows Scheduled Task script generation for OpenClaw. Environment values were written into gateway.cmd using unquoted set KEY=VALUE, which allowed Windows shell metacharacters in config-provided environment variables to break out of assignment context.

A command injection vulnerability existed in Windows Scheduled Task script generation for OpenClaw. Environment values were written into gateway.cmd using unquoted set KEY=VALUE, which allowed Windows shell metacharacters in config-provided environment variables to break out of assignment context.

In openclaw@2026.2.25, BlueBubbles group authorization could incorrectly treat DM pairing-store identities as group allowlist identities when dmPolicy=pairing and groupPolicy=allowlist. A sender that was only DM-paired (not explicitly present in groupAllowFrom) could pass group sender checks for message and reaction ingress. Per OpenClaw's SECURITY.md trust model, this is a constrained authorization-consistency issue, not a multi-tenant boundary bypass or host-privilege escalation.

In openclaw@2026.2.25, BlueBubbles group authorization could incorrectly treat DM pairing-store identities as group allowlist identities when dmPolicy=pairing and groupPolicy=allowlist. A sender that was only DM-paired (not explicitly present in groupAllowFrom) could pass group sender checks for message and reaction ingress. Per OpenClaw's SECURITY.md trust model, this is a constrained authorization-consistency issue, not a multi-tenant boundary bypass or host-privilege escalation.

OpenClaw exec allowlist/safeBins policy could be bypassed with attached short-option payloads (for example sort -o/tmp/poc), enabling file-write operations while still satisfying safeBins checks.

OpenClaw exec allowlist/safeBins policy could be bypassed with attached short-option payloads (for example sort -o/tmp/poc), enabling file-write operations while still satisfying safeBins checks.

DM pairing-store identities were incorrectly eligible for group allowlist authorization checks, enabling cross-context authorization in group message paths.

DM pairing-store identities were incorrectly eligible for group allowlist authorization checks, enabling cross-context authorization in group message paths.

assertBrowserNavigationAllowed() validated only http:/https: network targets but implicitly allowed other schemes. An authenticated gateway user could navigate browser sessions to file:// URLs and read local files via browser snapshot/extraction flows.

assertBrowserNavigationAllowed() validated only http:/https: network targets but implicitly allowed other schemes. An authenticated gateway user could navigate browser sessions to file:// URLs and read local files via browser snapshot/extraction flows.

Remote iMessage attachment fetches used SCP with trust-on-first-use host-key behavior and accepted unvalidated remote host tokens. Before the fix: SCP used StrictHostKeyChecking=accept-new in the remote attachment path. channels.imessage.remoteHost was not validated as a strict SSH host token.

BASH_ENV / ENV startup-file injection could lead to unintended pre-command shell execution when attacker-controlled environment values were admitted and then inherited by host command execution paths.

Rich text cell content rendered via v-html without sanitization, enabling stored XSS.

User-controlled content in comments and rich text cells was rendered via v-html without sanitization, enabling stored XSS.

Comments rendered via v-html without sanitization, enabling stored XSS.

An authenticated user with Creator role can inject arbitrary SQL via the DATEADD formula's unit parameter.

A vulnerability was detected in PhialsBasement nmap-mcp-server up to bee6d23547d57ae02460022f7c78ac0893092e38. Affected by this issue is the function child_process.exec of the file src/index.ts of the component Nmap CLI Command Handler. The manipulation results in command injection. The attack may be performed from remote. This product utilizes a rolling release system for continuous delivery, and as such, version information for affected or updated releases is not disclosed. The patch is identified as …