Manic is a sophisticated Android malware active since at least February 2026, targeting users primarily in Ukraine and multiple European countries. It combines spyware, banking fraud, and remote control capabilities, targeting numerous banking, government/eID, payment, crypto wallet, messaging, and authenticator apps. The malware uses Android Accessibility services to capture sensitive inputs like PINs, passwords, and SMS codes via transparent overlays. Uniquely, it can exfiltrate data through nearby infected devices using Wi-Fi Direct or Bluetooth when direct internet access to its command-and-control server is unavailable, enabling multi-hop data relays. Infection vectors remain unknown, but the malware employs anti-analysis techniques and in-memory payload loading. Users are advised to avoid unofficial APK sources and restrict Accessibility permissions.
A server-side request forgery (SSRF) vulnerability exists in the vLLM batch runner component of Red Hat AI Inference 3.4.4 (spyre). An attacker who can control batch input JSON may exploit this flaw to force the server to make arbitrary HTTP/HTTPS requests, potentially accessing internal services such as cloud metadata endpoints or internal APIs. This could lead to information disclosure or further compromise of the host system. Currently, no effective mitigation or fix meeting Red Hat's criteria is available.
LMDeploy deserializes disaggregated-serving peer messages with pickle. The handle_zmq_recv coroutine in lmdeploy/pytorch/disagg/conn/engine_conn.py reads peer-to-peer cache-free requests with recv_pyobj(), which deserializes the received bytes with pickle.loads(), and the isinstance check against DistServeCacheFreeRequest runs only after deserialization has already completed. The peer that supplies those bytes is caller-controlled: p2p_connect passes remote_engine_endpoint_info.zmq_address from the request body to connect() on the ZMQ PULL socket, and the POST /distserve/p2p_initialize and /distserve/p2p_connect endpoints in lmdeploy/serve/openai/api_server.py apply no authentication unless the server is started with api_keys, which defaults to None. A remote attacker can direct an engine to pull from a ZMQ endpoint under their control and execute arbitrary code in the engine process. Deployments that do not enable disaggregated serving are not affected, because the receive loop is only started once the migration backend accepts the connection.
A vulnerability in langgraph-api prior to version 0.10.0 allowed incomplete authorization checks during run creation. Specifically, the run-creation path used an authorization event ('assistants.search') that could be unhandled in some custom authorization setups, leading to omission of owner constraints. This allowed unauthorized users to create runs referencing private assistants owned by others and receive sensitive assistant configuration data in the response. The issue is fixed in langgraph-api 0.10.0 by unifying authorization events to 'assistants.read' during run and cron creation.
A vulnerability in langgraph-api versions prior to 0.10.0 allows relative webhook targets to bypass authentication by routing requests internally via a loopback transport. This can enable unauthorized creation or modification of runs on threads owned by other users, exposing limited thread metadata. The issue is fixed in version 0.10.0 by disabling loopback webhook delivery by default.
CVE-2026-55648 is an integer overflow vulnerability in FreeRDP prior to version 3.27.0. It occurs in the function freerdp_image_copy_from_icon_data, which calculates the required buffer size for icon data using 32-bit arithmetic. A malicious RDP server can send icon dimensions that cause this calculation to wrap around, bypassing bounds checks and potentially leading to out-of-bounds reads of attacker-controlled data. This affects RemoteApp clients using the vulnerable library path, while the xfreerdp client has a mitigation. The issue is fixed in FreeRDP version 3.27.0.
CVE-2026-14970 is a high severity buffer overflow vulnerability affecting IBM AIX versions 7.2 and 7.3, as well as IBM PowerVM VIOS 4.1. The issue causes the NIM server process to crash during client registration due to a buffer copy operation without checking the size of the input.
MyBooks is an enhanced and easy-to-use personal ebook management web server also known as Talebook. In 3.41.2 and earlier, the AdminSettings.post handler in webserver/handlers/admin.py accepts SOCIAL_AUTH key names without validating quotes or newline characters, and SettingsLoader.dumpfile in webserver/loader.py concatenates those names into the generated Python source file auto.py without escaping them. An administrator can submit a crafted SOCIAL_AUTH key name that closes the settings dictionary and injects arbitrary Python statements. The application later executes those statements because SettingsLoader.loadfile imports auto.py as a module, and setting autoreload to true invokes restart_async so a process supervisor restarts the service and triggers the import. Successful exploitation executes commands with the privileges of the application service account and can disclose data, modify files, establish persistence, or disrupt the service. Related authorization and registration vulnerabilities can reduce the effective privilege requirement in a chained attack, but the standalone vulnerability requires administrator access. This issue is fixed in version 3.42.0.
MyBooks is anebook management web server also known as Talebook. In 3.41.2 and earlier, the SignUp.post handler for POST /api/user/sign_up in webserver/handlers/user.py does not enforce the ALLOW_REGISTER configuration flag, even though the frontend hides registration controls when the flag is false. An unauthenticated remote attacker can call the endpoint directly and create a valid account on an instance whose administrator disabled public registration. The process_auth_header function in webserver/handlers/base.py also does not verify the account's active flag, so the newly created and unactivated account can authenticate immediately and access user-level API functionality. The bypass defeats the intended account-creation policy and can supply the low-privilege account required by related authorization vulnerabilities. This issue is fixed in version 3.42.0.
MyBooks is an ebook management web server also known as Talebook. In 3.41.2 and earlier, the AdminSettings.post handler for POST /api/admin/settings in webserver/handlers/admin.py applies the auth decorator but does not check the self.admin_user property, unlike the corresponding GET handler. Any authenticated regular user can therefore overwrite server configuration values including SMTP credentials, OAuth client secrets, storage paths, security feature flags, and autoreload settings. The process_auth_header function in webserver/handlers/base.py also fails to verify the matched account's active flag, allowing a registered but unactivated account to authenticate and reach the vulnerable handler. Exploitation can disclose secrets through configuration access paths, sabotage application behavior, force service restarts, and supply the settings needed for related code-injection attacks. This issue is fixed in version 3.42.0.
The npm package 'gfff5' version 1.0.0 masquerades as a system binary configuration tool but performs covert surveillance. When executed, it silently installs Python 3.12 on Windows and uses Python scripts to capture keystrokes, screenshots, clipboard data, and UI information. The collected data is sent to a hardcoded remote server. The package employs stealth techniques to hide its malicious activities from the user.
The npm package nibra1 version 1.0.0 contains malicious code that installs a hidden Python payload on Windows systems. This payload captures user input and screen content, sending it to a remote server, which then controls the victim's keyboard input to execute arbitrary commands. The installation attempts to silently install Python 3.12 if not present, using stealth techniques to avoid user detection.
The npm package 'saaa9' version 1.0.0 masquerades as a system binary configuration tool but contains covert Windows surveillance functionality. Upon execution, it silently elevates privileges, installs Python if missing, and runs a Python script that captures clipboard data, screenshots, and extracts on-screen text. The collected data is exfiltrated to a hardcoded remote server. The package's user interface is deliberately hidden to avoid detection, and its metadata does not reflect its true malicious behavior.
The npm package syjoy version 1.0.0 masquerades as a system binary configuration tool but contains malicious Python code that harvests user data and enables remote control of the host. On first run, it silently installs Python 3.12 and executes a Python script that captures keystrokes, clipboard data, screenshots, and extracts text from other application windows. The stolen data is sent to a hardcoded remote server, which can respond with commands that are automatically typed into the active window, effectively allowing remote input injection. The package uses hidden windows and elevated privileges to evade detection. This behavior indicates the package is a covert tool likely intended for unauthorized data collection and remote manipulation rather than legitimate configuration management.
The npm package 'syboy' version 1.0.0 contains malicious code that, upon execution, silently elevates privileges and runs a hidden Python script. This script continuously captures clipboard data, screenshots, and UI accessibility information, sending it to a hardcoded external server. The package also silently installs Python 3.12.3 as a fallback runtime. The malicious activity is designed to remain hidden from the user and screen-sharing tools.
The npm package bqq1 version 1.0.0 masquerades as a system binary configuration tool but contains malicious code that installs Python 3.12 silently and runs a Python script that captures keystrokes, clipboard data, screenshots, and UI text. This data is sent to a remote server, which then sends commands back to the infected host to type text remotely controlled by an attacker. The package uses stealth techniques including transparent windows and hotkeys to hide its activity.
---
_-= Per source details. Do not edit below this line.=-_
## Source: amazon-inspector (6808b9ae619e6cf9fdb59b52305e25b8e2fdb87167fbd78fae0ca6c1c07a5f1a)
On import, deepface-weight spawns a daemon background thread that polls the installer's working directory for `data/telethon_market_userbot.session` for approximately 10 minutes and POSTs the file to a hardcoded webhook.site endpoint (https://webhook.site/d6ea9c5b-4a85-4e59-9397-2bb5f9407c87). Telethon session files contain live authentication material granting full access to the associated Telegram account. The exfiltration destination is bound to a variable literally named `evil_server_url` with a Russian-language comment identifying it as the attacker's server. The package name mimics the popular `deepface` ML library but ships no machine-learning code; author metadata is a placeholder (`asdqwdasdqwdasd`) with a disposable email at playboot.com.
## Source: kam193 (92f87cc2da145e4ab765c912a3a9cfb36ed6e8d21231fd511aa032edcd4b3e2b)
During import, package exfiltrates the sensitive file with the Telegram session token.
---
Category: MALICIOUS - The campaign has clearly malicious intent, like infostealers.
Campaign: 2026-08-deepface-weights
Reasons (based on the campaign):
- files-exfiltration
- target:telegram
---
_-= Per source details. Do not edit below this line.=-_
## Source: amazon-inspector (152e5cdb52328dccd4eb61462c930b56f456dbdf759f0ce3005da3a7ecf4b688)
reqcrypt presents itself as a general-purpose HTTP client, but its internal PayloadProcessor unconditionally inspects every HTTP response for JSON keys named 'payloaddd', 'payload_gz', and 'payload_zlib', base64/gzip/zlib-decodes the associated values, and passes the decoded bytes to exec() inside the caller's Python process. This decode-and-exec path is wired into _process_response, which is invoked from every public verb (get/post/put/delete/patch/download/upload), so any server the caller contacts—or any network attacker able to tamper with a response—can push arbitrary Python code into the caller's process by adding one of those keys to a JSON body. In-source comments label the sinks as 'Hidden execution', and the behavior is not disclosed in the package's advertised API. Effect on any program using this library is a remote code execution channel driven by attacker-controlled or attacker-influenced HTTP responses.
## Source: kam193 (556f2178ed2505024fd91bc1c9c20b59d78783c27badd166a220493e5184fb39)
The package contains a hidden backdoor. The promised functionality is an HTTP request library with some additional functions. On every usage, code secretly checks for the presence of specific fields in the response, and if they are found, their content is secretly executed.
---
Category: MALICIOUS - The campaign has clearly malicious intent, like infostealers.
Campaign: 2026-08-reqcrypt
Reasons (based on the campaign):
- backdoor
---
_-= Per source details. Do not edit below this line.=-_
## Source: amazon-inspector (9c1cf8a0f273b75a4cf2b66c9b9fc351023b9f77d61e82c74d74534424ce0558)
On import, deepface_weights starts a daemon thread that polls every 10 seconds for the file `data/telethon_market_userbot.session` in the current working directory. When found, it POSTs the session file together with the local `os.getlogin()` value to the hardcoded endpoint https://webhook.site/730d2d03-5c78-4e0a-88df-9d8466b7e8aa. A Telethon `.session` file holds authenticated Telegram credentials, so exfiltration enables full takeover of the associated Telegram account. Package metadata is placeholder (author email `rozuvu@example.com`, description `Minimal example Python package`) and the name resembles the unrelated `deepface` face-recognition library, but the package ships none of that functionality — the stealer is its only behavior. Source comments in Russian label the destination as the attacker's server.
## Source: kam193 (a9b6a1255b998e6497198b80ffff88e0612a2d839c6f1b9859664436f96e5fce)
During import, package exfiltrates the sensitive file with the Telegram session token.
---
Category: MALICIOUS - The campaign has clearly malicious intent, like infostealers.
Campaign: 2026-08-deepface-weights
Reasons (based on the campaign):
- files-exfiltration
- target:telegram
Lemur manages TLS certificate creation. Prior to 1.9.3, _validate_acme_url enforced ACME_DIRECTORY_HOST_ALLOWLIST when an authority was created, but PUT /api/1/authorities/ passed options to lemur/authorities/service.py without applying the same check. A user holding an authority role could replace the stored acme_url with an internal service or instance-metadata URL such as 169.254.169.254. The next issuance operation loaded that value and passed it to ClientV2.get_directory, causing an outbound request from the Lemur backend. This bypassed the creation-time mitigation for CVE-2026-55166 and could expose internal services or cloud metadata. The fix revalidates acme_url whenever authority options are updated. This issue is fixed in version 1.9.3.