Vulnerabilities
All
My stack
Known exploited
Public exploit
Critical (9+)
Likely exploited (EPSS ≥ 50%)
Changed, 24 h
New, 24 h
Sort:
risk
modified
published
CVSS
EPSS
| CVE | CVSS | EPSS | Vendor, product | Description | Modified |
|---|---|---|---|---|---|
| CVE-2025-15616 | 7.1 high | 2.0% | wazuh | Wazuh wazuh-agent and wazuh-manager versions 2.1.0 before 4.8.0 contain multiple shell injection and untrusted search path vulnerabilities that allow attackers to execute arbitrary commands through various components including logcollector configuration, maild SMTP server tags, and Kaspersky AR script parameters. Attackers can exploit these vulnerabilities by injecting malicious commands through configuration files, SMTP server settings, and custom flags to achieve remote code execution on affected systems. | 14 Oct |
| CVE-2026-46449 | 6.7 medium | 2.0% | no CPE yet | The Akamai Guardicore Platform Agent (GPA) before 7.4 for macOS is affected by a local privilege escalation vulnerability within the wapptunneld component. The vulnerability is caused by a lack of input validation when the service uses network interface names to execute system utilities, resulting in command injection. An unprivileged local user with permissions to modify network interface names can exploit this by renaming a network interface to include a malicious shell payload via System Settings. When the "Log out from your IdP" action is subsequently triggered, the wapptunneld service processes the poisoned interface name, leading to the execution of arbitrary commands with root privileges. | 16 Oct |
| CVE-2021-21290 | 5.5 medium | 2.0% | netty, debian, debian_linux, quarkus | Netty is an open-source, asynchronous event-driven network application framework for rapid development of maintainable high performance protocol servers & clients. In Netty before version 4.1.59.Final there is a vulnerability on Unix-like systems involving an insecure temp file. When netty's multipart decoders are used local information disclosure can occur via the local system temporary directory if temporary storing uploads on the disk is enabled. On unix-like systems, the temporary directory is shared between all user. As such, writing to this directory using APIs that do not explicitly set the file/directory permissions can lead to information disclosure. Of note, this does not impact modern MacOS Operating Systems. The method "File.createTempFile" on unix-like systems creates a random file, but, by default will create this file with the permissions "-rw-r--r--". Thus, if sensitive information is written to this file, other local users can read this information. This is the case in netty's "AbstractDiskHttpData" is vulnerable. This has been fixed in version 4.1.59.Final. As a workaround, one may specify your own "java.io.tmpdir" when you start the JVM or use "DefaultHttpDataFactory.setBaseDir(...)" to set the directory to something that is only readable by the current user. | 8 Oct |
| CVE-2026-66297 | 5.0 medium | 2.0% | livebook | Improper Neutralization of Special Elements used in an OS Command (OS Command Injection) vulnerability in livebook-dev livebook allows command injection into generated deployment setup commands. LivebookWeb.Hub.Teams.DeploymentGroupAgentComponent.docker_instructions/2 and LivebookWeb.Hub.Teams.DeploymentGroupAgentComponent.fly_instructions/4 in lib/livebook_web/live/hub/teams/deployment_group_agent_component.ex interpolate deployment group environment variable values into the generated Docker and Fly.io setup commands without shell escaping. The values originate from the deployment group configuration and reach the sinks through Livebook.Hubs.Dockerfile.online_docker_info/3. Both sinks place the value inside a double-quoted shell word, so a value containing a command substitution such as $(...) or backticks is evaluated by the shell without any need to break out of the quoting, and a literal double quote terminates the quoted word and allows arbitrary further tokens. The generated command is displayed in the Livebook web interface with a copy button, so a user who copies it and runs it without reviewing it first executes the injected commands on their own machine, under their own account. An attacker requires privileges sufficient to set deployment group environment variables, while the resulting code execution occurs on the machine of whoever runs the generated command. The Kubernetes instructions are not affected, because they render the same values into a YAML manifest with escaping rather than into a shell command. This issue affects livebook: from 0.13.0 before 0.18.7 and from 0.19.0 before 0.19.9. | Sat 24 Oct |
| CVE-2026-15495 | 2.1 low | 2.0% | no CPE yet | A vulnerability has been found in SonicCloudOrg sonic-agent up to 2.7.2. The affected element is an unknown function of the file AndroidWSServer.java of the component Android WebSocket Server. The manipulation of the argument path leads to os command injection. The attack can be initiated remotely. The exploit has been disclosed to the public and may be used. The vendor was contacted early about this disclosure but did not respond in any way. This vulnerability only affects products that are no longer supported by the maintainer. | Sun 25 Oct |
| CVE-2026-15496 | 2.1 low | 2.0% | no CPE yet | A vulnerability was found in SonicCloudOrg sonic-agent up to 2.7.2. The impacted element is the function evalIsFailed of the file sonic-agent/src/main/java/org/cloud/sonic/agent/tests/script/GroovyScriptImpl.java of the component Groovy Script Handler. The manipulation results in os command injection. The attack can be launched remotely. The exploit has been made public and could be used. The vendor was contacted early about this disclosure but did not respond in any way. This vulnerability only affects products that are no longer supported by the maintainer. | Sun 25 Oct |
| CVE-2026-19243 | 2.1 low | 2.0% | no CPE yet | A security vulnerability has been detected in HKUDS nanobot up to 0.2.1. Impacted is the function ExecTool._guard_command/ExecTool._spawn of the file nanobot/agent/tools/shell.py of the component Shell Allowlist Handler. Such manipulation leads to os command injection. The attack can be executed remotely. The exploit has been disclosed publicly and may be used. Upgrading to version 0.3.0 is recommended to address this issue. The name of the patch is 4562. It is advisable to upgrade the affected component. Multiple issues were reported to the project. They reacted with a high level of professionalism and kindness: "These five reports are variants of the same root cause: validation of shell commands containing multiple segments, wrappers, comments, or chained commands. The issue was fixed by validating every executable shell segment against the configured allowlist". | Fri 23 Oct |
| CVE-2026-65145 | 10.0 critical | 1.0% | no CPE yet | NVIDIA NeMo Agent Toolkit contains a vulnerability where a attacker could cause a missing authentication issue. This vulnerability might lead to code execution, escalation of privileges, data tampering, denial of service, and information disclosure. | Tue 27 Oct |
| CVE-2026-46412 | 10.0 critical | 1.0% | no CPE yet | @beproduct/nestjs-auth is a NestJS authentication module for BeProduct IDS (Identity Server) with OpenID Connect support. Between 2026-05-11 20:19 UTC and 22:56 UTC, an attacker used a compromised npm publish token to publish 18 malicious versions of `@beproduct/nestjs-auth` (0.1.2 through 0.1.19). The postinstall payload attempted to harvest npm tokens (from `~/.npmrc`); GitHub personal access tokens, OAuth tokens (`gho_*`), and Actions OIDC tokens; AWS credentials (from environment variables and `~/.aws/credentials`); HashiCorp Vault tokens; and other secrets present in environment variables. Version `0.1.20` is a clean republish from the original `0.1.1` source tree. Anyone who installed any version in the range `>=0.1.2 <=0.1.19` should remove the package and clean the npm cache; install the clean version; rotate every credential present in the install environment, including all npm publish tokens, all GitHub PATs and OAuth tokens, AWS access keys, HashiCorp Vault tokens, and any other secret that was in env vars or config files at install time; scan affected hosts for indicators of compromise and, if any are found, treat the host as compromised and reimage; and check committed repository history for unexpected additions in `.claude/` or `.vscode/` directories. The worm is known to commit `setup.mjs` + hook configs to PR branches via automated agent runtimes. | Sat 24 Oct |
| CVE-2026-67429 | 10.0 critical | 1.0% | no CPE yet | Flyto2 Core is an execution kernel for automation and AI-agent workflows. Prior to 2.26.6, image.download and related file-writing modules use caller-controlled output_dir instead of validate_path_with_env_config and its FLYTO_SANDBOX_DIR confinement, allowing attacker-controlled response bytes to be written to arbitrary filesystem paths the process can access. This issue is fixed in version 2.26.6. | Sat 24 Oct |
| CVE-2026-47367 | 9.9 critical | 1.0% | no CPE yet | A malicious actor with access to the network and low privileges could exploit an Improper Input Validation vulnerability found in UID Enterprise Agent to execute a Command Injection on the host device. | Mon 26 Oct |
| CVE-2026-62830 | 9.9 critical | 1.0% | microsoft, azure_sre_agent | Missing authorization in Azure SRE Agent allows an authorized attacker to elevate privileges over a network. | Sat 24 Oct |
| CVE-2026-109516 | 9.8 critical | 1.0% | guruteam, bot for telegram on woocommerce | The Bot for Telegram on WooCommerce plugin for WordPress is vulnerable to Authentication Bypass leading to Account Takeover in all versions up to, and including, 1.3.0. The `/wp-json/woo-telegram/v1/main/` REST endpoint is registered with `permission_callback => '__return_true'` and performs no Telegram webhook signature verification, allowing an unauthenticated attacker to POST a forged bot update with a `message.chat.id` set to any WordPress login name; `bftow_get_user_system_id()` resolves this attacker-supplied value to the targeted account via `get_user_by('login', $tgId)` with no ownership proof, and writes a login token computed as `md5($_SERVER['HTTP_USER_AGENT'] . time())` — both inputs are fully attacker-controlled or bounded within seconds using the response's own `Date` header. The resulting token is redeemable at the unauthenticated, capability-check-free, and nonce-less `?action=bftow_create_order&bftow_token=` endpoint, which calls `wp_set_auth_cookie()` unconditionally via `login_user_by_id`, issuing a persistent authenticated session cookie for the targeted account to the attacker. This makes it possible for unauthenticated attackers to log in as any WordPress user, including administrators, given only that user's username. | 14 h ago |
| CVE-2026-76934 | 9.8 critical | 1.0% | no CPE yet | Stack-based Buffer Overflow Vulnerability in JP1/Automatic Job Management System 3. This issue affects JP1/Automatic Job Management System 3 - Manager: from 11-00 before 13-50-31; JP1/Automatic Job Management System 3 - Agent: from 11-00 before 13-50-31. | Tue 27 Oct |
| CVE-2026-76939 | 9.8 critical | 1.0% | no CPE yet | Heap-based Buffer Overflow Vulnerability in JP1/Automatic Job Management System 3. This issue affects JP1/Automatic Job Management System 3 - Manager: from 11-00 before 13-50-31; JP1/Automatic Job Management System 3 - Agent: from 11-00 before 13-50-31. | Tue 27 Oct |
| CVE-2026-40288 | 9.8 critical | 1.0% | praison, praisonai, praisonaiagents | PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the workflow engine is vulnerable to arbitrary command and code execution through untrusted YAML files. When praisonai workflow run <file.yaml> loads a YAML file with type: job, the JobWorkflowExecutor in job_workflow.py processes steps that support run: (shell commands via subprocess.run()), script: (inline Python via exec()), and python: (arbitrary Python script execution)—all without any validation, sandboxing, or user confirmation. The affected code paths include action_run() in workflow.py and _exec_shell(), _exec_inline_python(), and _exec_python_script() in job_workflow.py. An attacker who can supply or influence a workflow YAML file (particularly in CI pipelines, shared repositories, or multi-tenant deployment environments) can achieve full arbitrary command execution on the host system, compromising the machine and any accessible data or credentials. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents. | Mon 26 Oct |
| CVE-2026-40351 | 9.8 critical | 1.0% | fastgpt | FastGPT is an AI Agent building platform. In versions prior to 4.14.9.5, the password-based login endpoint uses TypeScript type assertion without runtime validation, allowing an unauthenticated attacker to pass a MongoDB query operator object (e.g., {"$ne": ""}) as the password field. This NoSQL injection bypasses the password check, enabling login as any user including the root administrator. This issue has been fixed in version 4.14.9.5. | Mon 26 Oct |
| CVE-2026-41497 | 9.8 critical | 1.0% | praison, praisonai | PraisonAI is a multi-agent teams system. Prior to version 4.6.9, the fix for PraisonAI's MCP command handling does not add a command allowlist or argument validation to parse_mcp_command(), allowing arbitrary executables like bash, python, or /bin/sh with inline code execution flags to pass through to subprocess execution. This issue has been patched in version 4.6.9. | Mon 26 Oct |
| CVE-2026-13448 | 9.8 critical | 1.0% | langflow | IBM Langflow OSS 1.0.0 through 1.10.1 Lanflow OSS contains an unauthenticated remote code execution vulnerability in the public flow build endpoint ( /api/v1/build_public_tmp/{flow_id}/flow ). The vulnerability stems from an incomplete denylist in the validate_public_flow_no_code_execution() function that fails to block several code-execution agent components including OpenDsStarAgent, CodeActAgentSmolagents, and CSVAgent. | Sat 24 Oct |
| CVE-2026-47391 | 9.8 critical | 1.0% | no CPE yet | PraisonAI is a multi-agent teams system. Prior to version 4.6.40, PraisonAI's first-party A2A server example exposes an unauthenticated A2A JSON-RPC endpoint and registers a `calculate(expression)` tool implemented with Python `eval()`. The example also binds to `0.0.0.0`. A remote unauthenticated attacker can send `message/send` to `/a2a`; the request reaches `agent.chat()`, and a real LLM can invoke the registered `calculate` tool. In testing with `gemini/gemini-2.5-flash-lite`, this resulted in arbitrary Python execution in the server process, confirmed by creation of a marker file from an unauthenticated HTTP request. The issue affects deployments following the official A2A example or similar unauthenticated public A2A deployments with unsafe tools. The default unauthenticated A2A surface also exposes task history and task cancellation APIs, increasing confidentiality and integrity impact. Version 4.6.40 patches the issue. | Sat 24 Oct |
| CVE-2026-47393 | 9.8 critical | 1.0% | no CPE yet | PraisonAI is a multi-agent teams system. CVE-2026-44338 (GHSA-6rmh-7xcm-cpxj) documents that PraisonAI ships a code-generator (`praisonai.deploy.api.generate_api_server_code`) that emits a Flask API server with authentication disabled by default. Users who follow the documented quickstart (`praisonai deploy --type api`) get a server that binds to `0.0.0.0` per the recommended sample YAML, exposes `/chat` and `/agents` endpoints, runs `praisonai.run()` on user-supplied JSON input — LLM orchestration with the API key materials present in the process environment, and does not require any authentication. Versions prior to 4.6.40 still ship the generator with `auth_enabled` defaulting to `False`. The fix shape is opt-in via `APIConfig(auth_enabled=True, auth_token=...)`. Version 4.6.40 fixes the issue. | Sat 24 Oct |
| CVE-2026-46994 | 9.8 critical | 1.0% | oracle, enterprise_manager_base_platform | Vulnerability in the Oracle Enterprise Manager Base Platform product of Oracle Enterprise Manager (component: Agent Next Gen). Supported versions that are affected are 13.5 and 24.1. Easily exploitable vulnerability allows unauthenticated attacker with network access via HTTPS to compromise Oracle Enterprise Manager Base Platform. Successful attacks of this vulnerability can result in takeover of Oracle Enterprise Manager Base Platform. CVSS 3.1 Base Score 9.8 (Confidentiality, Integrity and Availability impacts). CVSS Vector: (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H). | Sat 24 Oct |
| CVE-2026-70905 | 9.8 critical | 1.0% | oracle, access_manager | Vulnerability in the Oracle Access Manager product of Oracle Fusion Middleware (component: Agent infrastructure). Supported versions that are affected are 12.2.1.4.0 and 14.1.2.1.0. Easily exploitable vulnerability allows unauthenticated attacker with network access via SAML to compromise Oracle Access Manager. Successful attacks of this vulnerability can result in takeover of Oracle Access Manager. CVSS 3.1 Base Score 9.8 (Confidentiality, Integrity and Availability impacts). CVSS Vector: (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H). | 22 Oct |
| CVE-2026-45311 | 9.6 critical | 1.0% | no CPE yet | CodeWhale is a DeepSeek + MiMo coding agent in terminal. From 0.3.0 to 0.8.23, the run_tests tool executes cargo test in the workspace with ApprovalRequirement::Auto, meaning it runs without any user approval prompt. cargo test compiles and executes arbitrary code: test binaries, build.rs build scripts, and proc macros. While auto-approving test execution is a deliberate design choice, it creates an inconsistency in the security boundary. However, in a malicious repository, test code can execute arbitrary shell commands, exfiltrate credentials, or establish persistence with zero approval. The attack is amplified by AGENTS.md (auto-loaded into the system prompt), which can instruct the model to run tests proactively at session start. This vulnerability is fixed in 0.8.23. | Mon 26 Oct |
| CVE-2026-70477 | 9.5 critical | 1.0% | flowiseai, flowise | Flowise is a drag & drop user interface to build a customized large language model flow. Prior to 3.1.3, a prompt injection sent to a chatflow using a CSV Agent node can cause the LLM to respond with a malicious Python script that bypasses the blocklist validator and executes in an unsandboxed Pyodide environment. The specific flaw exists within the run method of the CSV_Agents class, where untrusted data is used to construct an LLM prompt and the resulting pythonCode is validated by validatePythonCodeForDataFrame before execution. An attacker can leverage this to execute arbitrary code in the context of the service account. This issue is fixed in 3.1.3. | Sat 24 Oct |
| CVE-2025-55108 | 9.5 critical | 1.0% | no CPE yet | The Control-M/Agent is vulnerable to unauthenticated remote code execution, arbitrary file read and write and similar unauthorized actions when mutual SSL/TLS authentication is not enabled (i.e. in the default configuration). NOTE: * The vendor believes that this vulnerability only occurs when documented security best practices are not followed. BMC has always strongly recommended to use security best practices such as configuring SSL/TLS between Control-M Server and Agent. * The vendor notifies that Control-M/Agent is not impacted in Control-M SaaS | 18 Oct |
| CVE-2026-74040 | 9.4 critical | 1.0% | no CPE yet | Wazuh versions from 3.9.0 before 4.14.8 contain an arbitrary file overwrite vulnerability in MasterHandler.process_files_from_worker allowing a cluster worker to overwrite protected master files. Attackers controlling a worker can use a trailing-slash basename bypass and missing extra_valid enforcement to overwrite ossec.conf and agent.conf, achieving root remote code execution across the manager and every agent. | Mon 26 Oct |
| CVE-2026-15265 | 9.4 critical | 1.0% | tenable, nessus_agent | A path traversal vulnerability in Tenable Agent 11.2.0 and 11.1.3 and lower allows a privileged attacker to write arbitrary files outside the intended plugin directory, potentially leading to remote code execution. | Sat 24 Oct |
| CVE-2026-50561 | 9.4 critical | 1.0% | no CPE yet | Yuxi is a large-model-based intelligent knowledge base and knowledge graph agent development platform. Prior to version 0.6.2, the project's authentication mechanism contains a flaw. In affected versions, the system does not sufficiently validate the identity token in the Authorization header — only performing a validity check. This allows an administrator token generated in another deployment instance or local testing environment to be used to access the backend management interfaces of a different affected instance. An attacker who obtains or constructs an acceptable administrator Authorization token may bypass normal login authentication and gain administrator privileges. This vulnerability could allow an attacker to access system configurations, invoke backend management APIs, create administrator accounts, and ultimately take over the system backend. This issue has been fixed in version 0.6.2. Before upgrading, users are advised to implement the following temporary measures: Set the environment variable `JWT_SECRET_KEY` to a non-default value, and configure a unique, sufficiently strong JWT/authentication key for each deployment instance; and/or avoid exposing backend management interfaces directly to the public network. | Sat 24 Oct |
| CVE-2026-69256 | 9.4 critical | 1.0% | flowiseai, flowise | Flowise is a drag & drop user interface to build a customized large language model flow. Prior to 3.1.3, the CSVAgent node allowed users to provide Python code that is executed through pyodide; although a denylist blocked dangerous Python constructs, pandas.read_pickle() could deserialize a pickled payload and achieve code execution without matching the denied words. The affected file is flowise-components/nodes/agents/CSVAgent/CSVAgent.ts, where user-supplied customReadCSVFunc is evaluated as pd.${customReadCSVFunc}. An authenticated user who can create or modify a chatflow can add a CSV Agent, place a malicious read_pickle payload in the Additional Parameters, save the chatflow, and trigger /api/v1/prediction/<UUID> to execute commands. This issue is fixed in version 3.1.3. | Sat 24 Oct |
| CVE-2026-69264 | 9.4 critical | 1.0% | flowiseai, flowise | Prior to 3.1.3, Flowise CSVAgent interpolates an attacker-controlled segment of the csvFile data URI directly into a Python source-code template that is then executed by Pyodide. Because Pyodide is loaded with the default js bridge to globalThis, which on Node.js exposes eval and dynamic import, the attacker can break out of the Python string literal, hand a JavaScript string to js.eval, dynamically import Node built-in modules such as fs and child_process, and execute arbitrary file I/O or OS commands as the Flowise process. The two validator paths around this code, validatePythonCodeForDataFrame and validateCustomReadCSVFunction, are never applied to the bootstrap template. A workspace user with chatflows:create or agentflows/chatflows update permission can plant a CSV Agent node with a crafted csvFile; once the chatflow is exposed via POST /api/v1/prediction/:id, any unauthenticated request triggers host remote code execution. This issue is fixed in version 3.1.3. | Sat 24 Oct |
| CVE-2026-112346 | 9.3 critical | 1.0% | no CPE yet | OpenHands Software Agent SDK before 1.52.0 contains an unauthenticated remote code execution vulnerability where the DockerWorkspace and Apptainer workspace launchers bind the sandbox agent server to all network interfaces (0.0.0.0) with an empty session_api_keys list, disabling authentication by default. Attackers who can reach the published container port can send unauthenticated HTTP and WebSocket requests to execute arbitrary shell commands as root in the sandbox, access the mounted workspace, and read plaintext secrets from the secrets store. | Mon 26 Oct |
| CVE-2026-77930 | 9.3 critical | 1.0% | no CPE yet | EvoScientist 0.1.0 before 0.2.5 contains an approval bypass vulnerability that allows attackers to run host shell commands by delegating them to async sub-agents lacking an interrupt_on gate. Attackers can plant instructions in ingested content such as fetched web pages, so the agent calls start_async_task and runs commands with user privileges and environment secrets. | 22 Oct |
| CVE-2026-109444 | 9.3 critical | 1.0% | no CPE yet | OpenSRE through 0.1.2026.10.19 contains an approval gate bypass vulnerability that allows attackers to execute shell_run commands without confirmation because the headless presenter's execution_allowed() always returns True. Attackers can inject instructions into alerts or Slack messages triaged by the agent to run commands on the host and exfiltrate LLM, Slack and cloud credentials. | 22 Oct |
| CVE-2026-109442 | 9.3 critical | 1.0% | no CPE yet | piflow 2.0.0 through 2.0.1 contains a remote code execution vulnerability in its DeepAgents chat agent, whose POST /chat and POST /chat/stream endpoints lack authentication. Attackers can instruct the agent to call the exec_shell tool, which runs arbitrary commands via subprocess.run without human approval or an allow-list, exposing environment secrets such as LLM API keys and database credentials. | 20 Oct |
| CVE-2026-77932 | 9.3 critical | 1.0% | no CPE yet | DB-GPT through 0.8.2 contains an unauthenticated remote code execution vulnerability because the /api/v1/chat/react-agent endpoint relies on a mock get_user_from_headers function that grants every caller the admin role. Unauthenticated attackers can send prompts that make the ReAct agent's unsandboxed code_interpreter and shell_interpreter tools run arbitrary Python or bash as root in the webserver container, including reaching the bundled database with default credentials. | 20 Oct |
| CVE-2026-110631 | 9.3 critical | 1.0% | no CPE yet | Komodo through 2.3.3 contains an authentication bypass vulnerability in the Periphery agent that allows unauthenticated attackers to connect as a trusted Core because empty core_public_keys and passkeys lists fail open by default. Attackers reaching port 8120 can present any generated keypair to issue Periphery API requests, including terminal sessions and container operations, executing commands on managed hosts typically as root. | 20 Oct |
| CVE-2026-103091 | 9.3 critical | 1.0% | no CPE yet | Missing Authorization (CWE-862) and Initialization of a Resource with an Insecure Default (CWE-1188) in the MCP proxy (HandleStreamPassThrough in mcp_service.go) and OIDC registration flow (GET /info and POST /register in internal/controlplane/server.go) in Google OSS Agent Mesh (SAM) versions 0.1.0-alpha.1 through 0.1.0-alpha.8 allow a remote attacker to self-enroll into the mesh and invoke arbitrary tools across all connected MCP backend nodes. Because GET /info publicly discloses OIDC configuration, POST /register auto-enrolls any valid OIDC token holder without claim filtering or ban checks, and the default policy binds sam:system:authenticated to wildcard services and targets, an attacker can obtain a mesh credential with full read, write, and execution access on backend services. Users should upgrade to version 0.1.0-alpha.9 or later, which enforces a fail-closed default policy, explicit policy-binding requirements on enrollment, and ban checks. | 19 Oct |
| CVE-2026-111322 | 9.3 critical | 1.0% | no CPE yet | Archon 0.2.13 through 0.11.1 contains a missing authentication vulnerability that allows unauthenticated attackers to access the /api/* control API, which binds to 0.0.0.0 by default. Attackers with network access can send messages to the AI coding agent, which runs with bypassPermissions, to execute commands on the host as the Archon user. | 16 Oct |
| CVE-2025-68952 | 9.3 critical | 1.0% | eigent | Eigent is a multi-agent Workforce. In version 0.0.60, a 1-click Remote Code Execution (RCE) vulnerability has been identified in Eigent. This vulnerability allows an attacker to execute arbitrary code on the victim's machine or server through a specific interaction (1-click). This issue has been patched in version 0.0.61. | 13 Oct |
| CVE-2026-41265 | 9.2 critical | 1.0% | flowiseai, flowise | Flowise is a drag & drop user interface to build a customized large language model flow. Prior to 3.1.0, the specific flaw exists within the run method of the Airtable_Agents class. The issue results from the lack of proper sandboxing when evaluating an LLM generated python script. Using prompt injection techniques, an unauthenticated attacker with the ability to send prompts to a chatflow using the Airtable Agent node may convince an LLM to respond with a malicious python script that executes attacker controlled commands on the flowise server. This vulnerability is fixed in 3.1.0. | Mon 26 Oct |
| CVE-2026-58370 | 9.2 critical | 1.0% | no CPE yet | Woodpecker before 3.15.0 matches the ApprovalAllowedUsers bypass list against pipeline.Author. For the GitLab forge driver, pipeline.Author is populated from the git commit author name (commit.author.name) carried in the webhook payload, which is attacker-controlled and not verified by GitLab. A user who can open a merge request from a fork can set the commit author name to match an entry in ApprovalAllowedUsers, causing needsApproval to return false so the pipeline runs without the required approval. This defeats the fork-approval security boundary and allows execution of attacker-controlled pipeline steps on a Woodpecker agent and exfiltration of CI secrets exposed to the run. Other built-in forge drivers (Gitea, Forgejo, GitHub, Bitbucket) derive pipeline.Author from the forge-validated sender/actor identity and are not affected. | Sun 25 Oct |
| CVE-2026-96620 | 9.2 critical | 1.0% | no CPE yet | Claude Agent SDK is a Python SDK for Claude Code. Prior to 0.2.121, the resume option is passed to the Claude CLI as a separate token without validation, so a dash-leading value is parsed as an independent CLI flag. An attacker can inject the --mcp-config flag to define a stdio MCP server, causing arbitrary command execution at CLI startup outside the SDK permission controls. The issue is exploitable when an application passes untrusted input as the session ID to resume. This issue is fixed in version 0.2.121. | 21 Oct |
| CVE-2026-109437 | 9.2 critical | 1.0% | no CPE yet | Cognition before 0.13.0 contains an unauthenticated remote code execution vulnerability due to defaulting its agent sandbox backend to local execution and shipping no authentication on its REST/SSE API. Attackers can create a session via POST /sessions and post a message via POST /sessions/{id}/messages without credentials, causing the Deep Agents ReAct loop to run attacker-directed shell commands through the local shell backend, resulting in remote code execution as the server process. | 20 Oct |
| CVE-2026-110616 | 9.2 critical | 1.0% | no CPE yet | NullHub through 2026.5.29 contains a missing authentication vulnerability that allows unauthenticated attackers to access the /api/* management API because the auth_token field defaults to null and is never assigned. Attackers sending requests without Authorization or Origin headers to non-loopback listeners can control agent instances, rewrite configurations including provider credentials, and manage cron jobs and MCP servers. | 20 Oct |
| CVE-2026-105772 | 9.1 critical | 1.0% | no CPE yet | Path Traversal (CWE-22) in FileArtifactService in Google ADK TypeScript (@google/adk) versions 0.4.0 before 0.6.0 allows a remote unauthenticated attacker to read files matching the artifact directory layout or write files (when an agent tool invokes saveArtifact) outside the intended artifact storage root via crafted relative or percent-encoded path traversal sequences (such as ../ or ..%2F) in the userId or sessionId parameters when the ADK API server is exposed with the non-default FileArtifactService enabled. Applications using the default InMemoryArtifactService are unaffected. Upgrade to @google/adk version 0.6.0 or later. | 20 Oct |
| CVE-2025-30201 | 9.1 critical | 1.0% | wazuh | Wazuh is a free and open source platform used for threat prevention, detection, and response. Prior to version 4.13.0, a vulnerability in Wazuh Agent allows authenticated attackers to force NTLM authentication through malicious UNC paths in various agent configuration settings, potentially leading NTLM relay attacks that would result privilege escalation and remote code execution. This issue has been patched in version 4.13.0. | 19 Oct |
| CVE-2026-73485 | 9.0 critical | 1.0% | flowiseai, flowise | Flowise before 3.1.3 contains a code injection vulnerability in the Airtable Agent node that allows unauthenticated attackers to execute arbitrary Python code by bypassing the pythonCodeValidator blocklist through obfuscation techniques. Attackers can send crafted prompts to a chatflow using the Airtable Agent node to inject malicious Python code that executes in an unsandboxed pyodide environment with full access to the host operating system. | Sat 24 Oct |
| CVE-2026-73486 | 9.0 critical | 1.0% | flowiseai, flowise | Flowise before 3.1.3 contains a code injection vulnerability in the CSV Agent node's customReadCSV parameter that allows authenticated attackers to execute arbitrary Python code. The validator uses a static regex blocklist that can be bypassed through obfuscation techniques, enabling attackers to execute code in the unsandboxed pyodide environment with full system access. | Sat 24 Oct |
| CVE-2026-73487 | 9.0 critical | 1.0% | flowiseai, flowise | Flowise before 3.1.3 contains a regex-based Python code validator bypass in CSV and Airtable Agent nodes that allows unauthenticated attackers to inject malicious code via prompt injection. Attackers can exploit unblocked pandas functions like pd.read_json() to exfiltrate datasets, perform SSRF against internal services, or achieve code execution through the unauthenticated prediction API. | Sat 24 Oct |
Download these 793 as CSV · subscribe to this list.
EPSS scores: 2026-10-29. NVD cursor: 2026-10-29 00:22. The News page shows only notable CVEs (KEV, your stack, new criticals, EPSS over 50%, or active exploitation per SSVC); this table has all of them.