QueryWeaver contains an authentication bypass vulnerability that allows unauthenticated attackers to obtain valid session tokens for existing accounts by submitting a signup request with a known victim email address. The signup route unconditionally creates and links a new token to the matching Identity via a Cypher MERGE operation before checking whether the email belongs to an existing account, causing the server to return a valid authenticated session token for the victim's identity without requiring any prior credentials or user interaction.

A stored cross-site scripting (XSS) vulnerability exists in the `POST /api/prompts/share` endpoint of parisneo/lollms (latest version). The endpoint stores attacker-controlled `prompt_content` into `DBDirectMessage.content` without server-side sanitization. When a victim opens the direct message (DM) thread, the message is rendered by the DM UI through `MessageContentRenderer`, which uses `v-html` to insert rendered HTML into the DOM. The frontend sanitizer, which is regex-based, fails to comprehensively sanitize attacker-controlled HTML, allowing malicious payloads to execute in the victim's browser context. This vulnerability enables any authenticated user to send a malicious prompt-share message to another user's inbox, leading to arbitrary JavaScript execution, authenticated actions as the victim, exposure of same-origin application data, and potential account takeover.

The urwid web display backend (urwid/display/web.py) generates web session identifiers (urwid_id) in Screen.start() by concatenating two random.randrange(10**9) calls that use Python's Mersenne Twister PRNG, which is not cryptographically secure. Each call consumes approximately 30 bits of PRNG state, and the Mersenne Twister internal state is approximately 19,937 bits, so an attacker who observes approximately 334 session IDs (for example via the X-Urwid-ID HTTP response header) can fully reconstruct the internal state and predict all past and future session IDs (Path B). The same identifier is also used as the filename of a FIFO created in the world-listable /tmp directory (for example /tmp/urwid375487765176907690.in), so any local user on the host can list /tmp to enumerate active session tokens directly (Path A). With a valid session ID, an attacker can read the victim's terminal screen via the polling endpoint, inject keystrokes into the victim's session (yielding OS-level code execution with the session owner's privileges if the session runs a shell), and inject exit sequences or flood the FIFO to terminate or crash the session. A prior Bandit S311 warning on this usage was suppressed with # noqa: S311 rather than fixed

OpenPLC_v3 contains a heap-based buffer overflow in the getData() function in webserver/core/modbus_master.cpp. getData() reads characters between two delimiters into a caller-supplied buffer with no size parameter and no bounds check. In parseConfig() the function is invoked with the 100-byte heap-allocated MB_device.dev_name field. An authenticated attacker with access to the OpenPLC web interface can send a crafted HTTP POST to the /modbus endpoint with an oversized device_name value; the value is persisted to mbconfig.cfg and parsed on load, overflowing dev_name and overwriting adjacent struct fields (protocol at offset 108, dev_address at offset 109, ip_port at offset 210). A 200-byte payload writes 100 bytes past the allocation. The result is heap corruption leading to runtime crash and denial of service of the PLC process control loop, with attacker-controlled overwrite of adjacent configuration fields. The upstream repository was archived on 2026-04-04 and no fix is expected; the vendor has confirmed the issue does not affect OpenPLC Runtime v4.

SurrealDB versions before 1.1.0 fail to properly parse the ID, DB, and NS headers in HTTP REST API requests containing special characters. Unauthenticated attackers can send crafted HTTP requests with malformed header values to trigger an uncaught exception that crashes the server.

SurrealDB before 1.1.1 contains a format string vulnerability in the rquickjs Exception::throw_type function when scripting is enabled. Attackers with scripting privileges can supply format string sequences in error inputs to read arbitrary memory or execute code with SurrealDB process privileges.

SurrealDB before 1.5.5 (and 2.0.0-beta before 2.0.0-beta.3) accepts an arbitrary object in the signin and signup operations of the RPC API without recursively validating it for non-computed values. When a record access method defines a SIGNIN or SIGNUP query and the RPC API is exposed to untrusted users, an unauthenticated attacker can encode a binary object containing a subquery using the bincode serialization format and supply it in place of credentials. The subquery is then executed within the database owner's SIGNIN/SIGNUP query under a system user session with the editor role, allowing the attacker to select, create, update, and delete non-IAM resources (though not view the query results directly, and not affect IAM resources, which require the owner role).

SurrealDB before 1.0.1 sets default table permissions to FULL instead of NONE, allowing SELECT, CREATE, UPDATE, and DELETE operations on tables without explicit permissions. Attackers with database access or unauthenticated users on publicly exposed instances can perform unrestricted operations on unprotected tables within their authorization scope.

Impact: @fastify/http-proxy versions from 9.4.0 up to and including 11.5.0 fail to validate the resolved WebSocket destination path against the configured rewrite prefix. The WebSocket routing path in WebSocketProxy.findUpstream resolves the destination via the WHATWG URL constructor, which collapses dot segments, so a crafted upgrade request with path traversal sequences can escape the rewrite prefix and reach upstream endpoints that were not meant to be exposed by the proxy. This is a variant of CVE-2021-21322 in a code path that never went through the HTTP fix in fastify/reply-from. Exploitation requires a non-normalizing WebSocket client, since browsers and the ws package normalize the request path before sending, but raw HTTP clients or downstream proxies that forward the request target unchanged make the attack reachable in production topologies. Patches: upgrade to @fastify/http-proxy 11.6.0. Workarounds: none.

uproot dynamically generates Python class source code from ROOT TStreamerInfo records in a file and compiles it at runtime. Some file-controlled streamer metadata fields (for example, streamer element names) are interpolated into the generated Python source without safe quoting via repr() or the !r format specifier. An attacker who can supply a crafted ROOT file can place Python expression-breaking content into a streamer metadata field. When uproot generates and invokes the corresponding reader method, the injected Python expression is evaluated in the context of the process opening the file, resulting in arbitrary Python code execution in applications that open or process attacker-controlled ROOT files with affected uproot code paths.

Impact: @fastify/reply-from versions from 8.3.1 up to but not including 12.6.4 build the internal URL cache key by concatenating the destination and source path without a delimiter. Different destination and source pairs can therefore produce the same key while resolving to different upstream URLs. When getUpstream selects an upstream from request data, a URL cached for one upstream can be reused for a request intended for another upstream, causing cross-upstream data access and modification. The default configuration is affected. Setting disableCache to true prevents the behavior. Patches: upgrade to @fastify/reply-from 12.6.4. Workarounds: pass disableCache: true when registering the plugin.

A vulnerability was found in Shibby Tomato 1.28. This vulnerability affects the function sub_42537C of the component Scheduler Name Handler. The manipulation of the argument a1 results in stack-based buffer overflow. It is possible to launch the attack remotely. This project is superseded by FreshTomato.

A vulnerability has been found in Shibby Tomato 1.28 RT-N5x MIPSR2 Build 124. This affects the function sub_40BB50 of the file /proc/webmon_recent_domains. The manipulation leads to stack-based buffer overflow. It is possible to initiate the attack remotely. This project is superseded by FreshTomato.

A flaw has been found in Shibby Tomato 1.28 RT-N5x MIPSR2 Build 124. Affected by this issue is the function setup_conntrack of the file /sbin/rc. Executing a manipulation of the argument ct_tcp_timeout can lead to out-of-bounds write. The attack may be performed from remote. This project is superseded by FreshTomato.

VMware Avi Load Balancer contains a directory traversal vulnerability. Flaws in file path validation allow malicious, authenticated network users to perform directory traversal attacks. Affected versions: 32.1.1 (fixed in 32.1.2) 31.1.1 through 31.2.2 (fixed in 31.2.2-2p3) 30.1.1 through 30.2.6 (fixed in 30.2.7) 22.1.1 through 22.1.7 (fixed in 30.2.7)

VMware Avi Load Balancer contains a remote code execution vulnerability. A malicious authenticated user with network access may be able to inject and execute code. Affected versions: 32.1.1 (fixed in 32.1.2) 31.1.1 through 31.2.2 (fixed in 31.2.2-2p3) 30.1.1 through 30.2.6 (fixed in 30.2.7) 22.1.1 through 22.1.7 (fixed in 30.2.7)

VMware Avi Load Balancer contains a remote code execution vulnerability. A malicious user with network access may be able to access the Avi Control plane and execute code remotely. Affected versions: 32.1.1 (fixed in 32.1.2) 31.1.1 through 31.2.2 (fixed in 31.2.2-2p3) 30.1.1 through 30.2.6 (fixed in 30.2.7) 22.1.1 through 22.1.7 (fixed in 30.2.7)

VMware Avi Load Balancer contains an authorization bypass vulnerability. A malicious actor on the network can access a limited subset of the Avi Control Plane without proper authorization. Affected versions: 32.1.1 (fixed in 32.1.2) 31.1.1 through 31.2.2 (fixed in 31.2.2-2p3) 30.1.1 through 30.2.6 (fixed in 30.2.7) 22.1.1 through 22.1.7 (fixed in 30.2.7)

view_component is a framework for building reusable, testable, and encapsulated view components in Ruby on Rails. From 4.0.0 until 4.12.0, ViewComponent::Base#around_render can return HTML-unsafe strings that bypass the escaping behavior applied to normal #call return values. This creates an XSS risk when downstream applications use around_render to wrap, replace, instrument, or conditionally return content that includes user-controlled data, and ViewComponent::Collection#render_in can amplify the issue by joining per-item results and marking the entire output html_safe, converting raw unsafe output into an ActiveSupport::SafeBuffer. This issue is fixed in version 4.12.0.

IBM Langflow OSS 1.0.0 through 1.10.1 can allow an authenticated attacker to exploit the SaveToFile component to read and modify another user's uploaded files by specifying absolute paths pointing to victim storage locations. In append mode, the attacker's workflow reads victim file contents, appends attacker-controlled data, and uploads a copy containing victim data to the attacker's namespace (confidentiality breach). In overwrite mode, the attacker can replace victim file contents with arbitrary data (integrity breach). This breaks the storage ownership boundary between users.

css_parser is a Ruby CSS parser. From 2.2.0 until 3.0.0, CssParser::Parser#read_remote_file in lib/css_parser/parser.rb, and therefore load_uri! and the @import-following branch of add_block!, issued HTTP and HTTPS requests against any host, port, and URI without a scheme allowlist, host or IP filtering, or protection against link-local, loopback, or RFC-1918 addresses. Location: redirects were followed recursively back into the same function, which also serviced file:// URIs, so a single attacker-controlled HTTP redirect could upgrade the bug from SSRF to arbitrary local file disclosure. Any consumer of css_parser that hands it attacker-influenced CSS together with a base_uri: option is exposed. This issue is fixed in version 3.0.0.

IBM Langflow OSS 1.0.0 through 1.10.1 Langflow could allow an authenticated user to execute arbitrary commands with elevated privileges on the system due to improper validation of user supplied input in the Python Interpreter component.

systeminformation is a System and OS information library for node.js. Prior to 5.31.7, networkInterfaces() on Linux is vulnerable to OS command injection through the Debian/Ubuntu interfaces(5) source directive because lib/network.js checkLinuxDCHPInterfaces() reads /etc/network/interfaces, extracts a source <path> token from file content, and interpolates it unquoted into cat ${file} 2> /dev/null | grep 'iface\|source' executed by execSync(cmd, util.execOptsLinux), allowing a path containing shell metacharacters to execute commands in any process that calls networkInterfaces(), including via getStaticData() and getAllData(). This issue is fixed in version 5.31.7.

IBM Langflow OSS 1.0.0 through 1.10.0 allows an authenticated attacker to create a malicious flow pointing to an attacker-controlled URL that returns a specially crafted Content-Disposition header (e.g., filename="../../../target/path" ), enabling arbitrary file write operations with attacker-controlled content to any path accessible by the Langflow process.

joserfc is a Python library that provides an implementation of several JSON Object Signing and Encryption (JOSE) standards. Prior to 1.6.8, joserfc.jwt.decode accepts attacker-forged HMAC-signed tokens when the caller-supplied verification key is the empty string or None, because HMACAlgorithm.sign and HMACAlgorithm.verify in src/joserfc/_rfc7518/jws_algs.py pass the output of OctKey.get_op_key(...) to hmac.new(...) and OctKey.import_key in src/joserfc/_rfc7518/oct_key.py only emits a SecurityWarning for keys shorter than 14 bytes without rejecting zero-length input. This issue is fixed in version 1.6.8.

IBM Langflow OSS 1.0.0 through 1.10.0 allows authenticated users to override component parameters at runtime via the API. A critical security flaw exists in the parameter filtering mechanism within the `apply_tweaks()` function.

Pimcore is an Open Source Data & Experience Management Platform. Prior to 11.5.17 (LTS) and 12.3.6, the columnConfigAction endpoint in bundles/CustomReportsBundle/src/Controller/Reports/CustomReportController.php passes malicious SQL configuration through CustomReportController:columnConfigAction, SqlAdapter::getColumns, SqlAdapter::buildQueryString, and Db::fetchAssociative(), allowing an attacker with the reports_config permission to use arbitrary SELECT queries, UNION statements, dangerous database functions, and error-based SQL injection to exfiltrate or manipulate database data. This issue is fixed in versions 11.5.17 (LTS) and 12.3.6.

Pimcore is an Open Source Data & Experience Management Platform. Prior to 11.5.17 (LTS) and 12.3.7, Pimcore's WebDAV asset endpoint exposes a MOVE operation through /asset/webdav{path} without an authentication plugin in bundles/CoreBundle/src/Controller/WebDavController.php, and models/Asset/WebDAV/Tree.php performs asset mutation and deletion through models/Asset.php before checking a current Pimcore user or the rename, delete, create, or publish permissions, allowing unauthorized asset deletion, moves, or overwrites. This issue is fixed in versions 11.5.17 (LTS) and 12.3.7.

Pimcore is an Open Source Data & Experience Management Platform. Prior to 11.5.17 (LTS) and 12.3.7, multiple Pimcore locations call PHP's unserialize() on data from database columns and filesystem files without the allowed_classes restriction, including lib/Tool/Authentication.php, models/Site/Dao.php, models/DataObject/ClassDefinition/CustomLayout/Dao.php, models/Tool/TmpStore/Dao.php, models/Asset/WebDAV/Service.php, and admin-ui-classic-bundle/src/Helper/Dashboard.php, enabling object injection and remote code execution if an attacker can control the serialized data source. This issue is fixed in versions 11.5.17 (LTS) and 12.3.7.

Agentic-Flow is an AI agent orchestration platform. Prior to 2.0.14, agentic-flow MCP server tools in src/mcp/standalone-stdio.ts, src/mcp/fastmcp/servers/claude-flow-sdk.ts, src/mcp/fastmcp/servers/stdio-full.ts, src/mcp/fastmcp/servers/http-streaming-updated.ts, src/mcp/fastmcp/servers/http-sse.ts, src/mcp/fastmcp/servers/poc-stdio.ts, src/mcp/fastmcp/tools/agent/{execute,list,parallel}.ts, src/mcp/fastmcp/tools/swarm/orchestrate.ts, and src/mcp/fastmcp/tools/hooks/pretrain.ts interpolated attacker-influenceable tool parameters such as agent, task, name, language, and agentdb directly into shell command strings passed to execSync(), allowing arbitrary OS command execution with the privileges of the MCP server user. This issue is fixed in version 2.0.14.

AsyncSSH is a Python package which provides an asynchronous client and server implementation of the SSHv2 protocol on top of the Python asyncio framework. Prior to 2.23.0, AsyncSSH expands the OpenSSH-compatible AuthorizedKeysFile %u token in asyncssh/config.py, asyncssh/connection.py, asyncssh/auth_keys.py, and asyncssh/misc.py with the raw SSH username during pre-authentication server config reload, allowing a server configured with AuthorizedKeysFile authorized_keys/%u to read an authorized-keys file outside the intended directory when the SSH username contains /, \, or .. path traversal segments and authenticate with an attacker-selected key file. This issue is fixed in version 2.23.0.

SCRAM (Salted Challenge Response Authentication Mechanism) is part of the family of Simple Authentication and Security Layer (SASL, RFC 4422) authentication mechanisms. Prior to 3.3, a flaw in com.ongres.scram:scram-client and com.ongres.scram:scram-common allows an attacker capable of a TLS man-in-the-middle attack to silently downgrade a connection from SCRAM-SHA-256-PLUS with channel binding to standard SCRAM-SHA-256 without channel binding when TlsServerEndpoint processes an X.509 certificate using a modern signature algorithm such as Ed25519; getChannelBindingData() can return an empty byte array after NoSuchAlgorithmException, and the ScramClient builder treats that as absent channel-binding data. This issue is fixed in version 3.3.

ForgeCode (tailcallhq/forgecode), an AI pair-programming CLI, automatically loads and executes the MCP servers defined in a repository's .mcp.json file on startup without user confirmation. A malicious repository can supply a crafted .mcp.json whose mcpServers entries specify arbitrary command and args values (for example, command: bash with args: ['-c', 'touch /tmp/pwned']). When a user runs the forge CLI inside a cloned untrusted repository, the specified commands are spawned with the invoking user's privileges, resulting in arbitrary code execution. This provides a reliable initial-access and persistence primitive against developers who evaluate untrusted repositories with ForgeCode.

An unauthenticated reflected cross-site scripting (XSS) vulnerability exists in Sangoma Switchvox SMB Edition version 8.3 (104997). The application fails to properly sanitize the portal parameter supplied to the invalid_browser and invalid_browser_login handlers. User-supplied data is reflected into JavaScript generated by the application, allowing attacker-controlled script execution within a victim's browser.

Open Event Server through 1.19.1 contains a missing authentication vulnerability that allows unauthenticated attackers to export the complete member roster of any group, including email addresses, names, join dates, and roles, by submitting requests to the group followers CSV export endpoint which lacks any authentication decorator. Attackers can enumerate sequential group IDs via brute-force, trigger an export via the unauthenticated POST endpoint, then poll the unauthenticated task status endpoint until completion to retrieve a download URL containing the full member CSV.

The Joomla extension ChronoForms is vulnerable to an unauthenticated stored XSS vulnerability.

A path traversal vulnerability was identified in GitHub Enterprise Server that allowed an attacker who had code execution inside the Dependabot updater container to write files to arbitrary repository paths, including GitHub Actions workflow files under .github/workflows/ as the path validation did not check the effective path which the attacker could control through the dependency file's directory and symlink target. If the repository used a pull_request_target workflow or had auto-merge enabled, an injected workflow could execute with access to the repository's GitHub Actions secrets. This vulnerability affected all versions of GitHub Enterprise Server prior to 3.22 and was fixed in versions 3.21.3, 3.20.5, 3.19.9, 3.18.12, 3.17.18.

Missing Authorization in Google Cloud Firebase Studio versions prior to 2026-04-15 on Google Cloud Platform allows an attacker to download other users' deployed source code and access sensitive data via unauthorized GCS URL signing requests. This vulnerability was patched on 15 April 2026, and no customer action is needed.

Cursor for Windows version 3.2.16 contains a binary planting vulnerability that allows remote attackers to achieve arbitrary code execution by placing a malicious git.exe file in the repository root directory. When a developer clones and opens a crafted repository, Cursor automatically resolves and executes the workspace-resident git.exe during IDE startup and on a recurring timed cadence without any user interaction, running the malicious binary under the privileges of the current user.

SigNoz through 0.133.0 contains an open redirect vulnerability in the SSO authentication flow that allows unauthenticated attackers to steal session tokens from any user on instances configured with Google OAuth, SAML, or OIDC. Attackers can call the unauthenticated sessions context endpoint with a ref parameter pointing to an attacker-controlled host, deliver the resulting crafted login URL to a victim, and receive the victim's access and refresh tokens when they complete SSO authentication.

Dancer::Plugin::Auth::Google versions through 0.07 for Perl have TLS verification disabled. The default user agent is initialised with SSL_verify_mode explicitly disabled. An attacker with network man-in-the-middle (MITM) capability between the Dancer application and googleapis.com can intercept the OAuth2 token exchange and userinfo fetch, return a forged access_token and user profile, and be logged in to the Dancer application as any Google user.

Improper Validation of Specified Quantity in Input in ZenHive mpp allows an unauthenticated remote client to drain the fee-payer wallet, resulting in denial of service for legitimate clients. When the mpp Elixir library is configured as fee payer (fee_payer: true), the MPP.Methods.Tempo payment method co-signs and broadcasts a client-supplied EVM transaction without first validating that the client-supplied gas_limit is sufficient to complete the intended call. A malicious client can submit a signed transferWithMemo transaction with gas_limit deliberately set just below the amount required for successful execution. The server co-signs the transaction and broadcasts it via rpc_broadcast_sync. The transaction runs out of gas during EVM execution and reverts, but the fee-payer wallet is still charged for the burned gas while the client pays nothing and receives no resource. Repeated requests from one or more malicious clients drain the fee-payer wallet at near-zero cost to the attacker, ultimately preventing the server from sponsoring gas for legitimate payment requests. The wait_for_confirmation = false (optimistic) path is also affected: it invokes simulate_payment_call via eth_call, but that simulation omits the gas parameter and therefore does not catch out-of-gas conditions. This issue affects mpp: from 0.2.0 before 0.6.0.

Improper Validation of Specified Quantity in Input in ZenHive mpp allows an unauthenticated remote client to inflate the fee-payer's gas cost per payment by a large multiplier, degrading the sponsor's operating margin. When the mpp Elixir library is configured as fee payer (fee_payer: true), MPP.Tempo.Transaction.cosign_fee_payer/3 re-signs the client-supplied base fields of the 0x76 AASigned envelope verbatim, including the EIP-2930 access list, without validating its length or contents. EIP-2930 access list entries incur intrinsic gas (~2,400 gas per address, plus 1,900 gas per storage key) charged before any opcode executes, regardless of whether the listed addresses are ever touched. A malicious client submits a valid transferWithMemo call alongside a large number of fabricated access-list entries. The server co-signs and broadcasts the transaction. The intended transfer executes normally, but the fee-payer wallet pays a large multiple of the expected gas cost with no corresponding on-chain work. At the maintainer's default of 137 access-list entries (fitting within Bandit's 10,000-byte per-header-field limit) and 100 Gwei max_fee_per_gas, per-payment gas cost rises from ~51,287 to ~380,087 gas, a 7.4x multiplier. Sustained abuse destroys the sponsor's operating margin on low-cost payments and, over time, drains the fee-payer wallet. This issue affects mpp: from 0.2.0 before 0.6.0.

Improper Validation of Specified Quantity in Input in ZenHive mpp allows an unauthenticated remote client to drain the fee-payer wallet in a single request by naming an arbitrarily high gas price. When the mpp Elixir library is configured as fee payer (fee_payer: true), MPP.Tempo.Transaction.cosign_fee_payer/3 re-signs the client-supplied base fields of the 0x76 AASigned envelope verbatim, including max_fee_per_gas and max_priority_fee_per_gas, without validating that they are within reasonable bounds. A malicious client embeds arbitrarily large values for these fields in the signed envelope. The server co-signs and broadcasts the transaction. The effective_gas_price billed against the fee-payer wallet is derived from the attacker-supplied ceilings, so the server pays those inflated per-gas rates out of its own wallet. A single crafted request can drain the wallet entirely, after which the server can no longer sponsor gas for legitimate payment requests. This issue affects mpp: from 0.2.0 before 0.6.0.

The User Registration & Membership WordPress plugin before 5.2.3 does not validate that the membership tier submitted during public registration is one of the tiers allowed by the registration form before assigning that tier's associated user role, allowing unauthenticated users to register into an arbitrary published membership tier and obtain its role — up to administrator when such a tier exists.

The Paid Membership Plugin, Ecommerce, User Registration Form, Login Form, User Profile & Restrict Content – ProfilePress plugin for WordPress is vulnerable to Arbitrary File Upload in all versions up to, and including, 4.16.18 via the allowed_mime_types function. This is due to the unconditional registration of an upload_mimes filter that adds executable file extensions (.exe, .apk, .msi) to the global WordPress MIME allowlist, without scoping the expansion to digital-product upload contexts. This makes it possible for authenticated attackers, with author-level access and above, to upload files that may be executable, which makes remote code execution possible. This filter is registered globally on every request regardless of whether the digital products feature is configured or in use, meaning the expanded MIME allowlist affects all WordPress upload contexts site-wide.

The Grav API plugin (getgrav/grav-plugin-api) before 1.0.0-rc.16 accepts JWT access tokens through the ?token= URL query parameter on every API route (JwtAuthenticator::extractBearerToken fallback). Because tokens are embedded in URLs, they are logged verbatim in web server access logs, leaked via the Referer header, stored in browser history, and captured by upstream proxy and CDN logs, exposing valid admin access tokens. A leaked token grants unauthorized API access, including reading configuration and user data, creating admin accounts, modifying system settings, and deleting pages.

Grav before 2.0.4 fails to restrict cURL protocols in webhook dispatch, allowing authenticated users with api.webhooks.write permission to create webhooks with file://, dict://, or gopher:// URLs. Attackers can trigger webhook events to read local files, access process information, or pivot to internal services via unrestricted protocol handlers.

grav-plugin-api before 1.0.6 fails to validate super-admin status in createApiKey, generate2fa, and disable2fa endpoints, allowing non-super api.users.write managers to escalate to super-admin. Attackers can mint API keys bound to super-admin accounts or strip 2FA from super-admin users to achieve full instance takeover.

The Grav API plugin (getgrav/grav-plugin-api) before 1.0.6 contains an authorization bypass: API keys can be created with a restricted scopes array, but the ApiKeyAuthenticator class never reads or enforces these scopes. It loads and returns the owning user's full account object, so a key created with limited scopes (e.g. read-only) can perform any write, delete, or administrative operation the owning user is authorized for. Fixed in 1.0.6.