PraisonAI's unauthenticated A2A official example can reach real LLM-driven `eval()` tool execution
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 de
PraisonAI spider_tools SSRF protection bypass via alternate loopback host encodings
PraisonAI is a multi-agent teams system. Prior to version 4.6.40 of PraisonAI, corresponding to version 1.6.40 of praisonaiagents, `spider_tools` URL validation can be bypassed using alternate loopback host encodings. The tool contains a URL validation function intended to block local or unsafe targets before fetching attacker-controlled URLs. However, the validation only blocks a small set of exact host strings such as `localhost` and `127.0.0.1`. It does not normalize hostnames, resolve DNS, parse numeric IPv4 variants, or validate the final resolved IP address before making the request. As a result, certain URLs may bypass the protection a
Shamefile has an arbitrary file read via shamefile.yaml in shame next
Shamefile is a linter for undocumented linter warnings. Prior to version 0.1.7, a path traversal vulnerability in `shame next` allows an attacker-controlled `shamefile.yaml` to disclose contents of files outside the repository, one line at a time, to the terminal of a user who runs the command. See patch commit for technical details. The issue is fixed in 0.1.7. Upgrade to either 0.1.7 or later versions to incorporate the patch. As a workaround, do not run `shame next` against untrusted `shamefile.yaml`. Use `shame me --dry-run` for CI validation.
Flask-Security-Too OAuth reauthentication freshness bypass via cross- user OAuth identity acceptance
Flask-Security-Too allows users to add security features to their Flask applicationa. Version 5.8.0's OAuth reauthentication flow can mark a session as fresh after verifying an OAuth account that belongs to a different user. If an attacker can operate an already-authenticated but stale victim session, they can complete OAuth verification using their own OAuth identity. The victim session is then treated as recently reauthenticated, allowing freshness-protected account actions to proceed. Version 5.8.1 contains a fix for this issue.
Unsafe Deserialization in keras.layers.TorchModuleWrapper.from_config
A vulnerability in keras-team/keras version 3.15.0 allows unsafe deserialization of attacker-controlled PyTorch pickle data through the public `keras.layers.TorchModuleWrapper.from_config` method. This method invokes `torch.load(..., weights_only=False)` without requiring an explicit unsafe opt-in, such as a `safe_mode=False` parameter. When called outside a `SafeModeScope(True)` context, the absence of an ambient safe mode state permits unsafe deserialization by default. This issue can lead to arbitrary code execution if untrusted Keras layer configurations are processed using this method. The vulnerability arises because the method does not
SurrealDB before 1.0.1 Insecure Default Table Permissions
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.
dd-trace-py: Improper parsing of W3C baggage headers may lead to DoS
Datadog dd-trace-py is the Datadog Python APM client. Prior to 4.8.2, Datadog tracing libraries that implement W3C baggage propagation parse incoming baggage HTTP headers without enforcing DD_TRACE_BAGGAGE_MAX_ITEMS or DD_TRACE_BAGGAGE_MAX_BYTES limits on the extract path. A remote, unauthenticated attacker can send a request whose baggage header contains an arbitrarily large number of comma-separated key-value pairs or a single very large value, causing unbounded CPU and memory consumption and enabling a remote denial of service against HTTP services with baggage propagation enabled. This issue is fixed in version 4.8.2.
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.
Crítica · CVSS 9.8EPSS 0,87%Código de explotación públicoCorrección disponible
Authentication Bypass in Webhook Endpoints Allowed Unauthorized Flow Execution
IBM Langflow OSS 1.0.0 through 1.10.0 has a vulnerability in Langflow's webhook authentication logic allows unauthenticated users to trigger the execution of any flow. The system incorrectly bypasses API key validation when the WEBHOOK_AUTH_ENABLE configuration is set to False (which is the default setting). This allows a remote attacker who knows a flow's UUID to execute it as if they were the owner, potentially leading to Remote Code Execution (RCE).
AsyncSSH `AuthorizedKeysFile %u` path traversal allows attacker-selected authorized keys to authenticate a traversal username
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
### Impact
A stored XSS affecting RichText fields. RichTextValue.output returns the raw, unsanitized stored value whenever the stored mimeType equals the outputMimeType. Because the safe-HTML output type (`text/x-html-safe`) is the type that signifies "already sanitized", any value whose stored mimeType equals it bypasses the safe_html transform entirely on render. The transform itself is sound — it correctly strips `on*` event-handler attributes and `javascript:/data:` URIs; the defect is that it is never invoked for these values. The unsanitized value is then emitted via `tal:content="structure ..."`, which performs no escaping, so the pay
Zeroconf: Unvalidated rdlength in record payload readers allows LAN-local cache corruption via crafted mDNS packet
Zeroconf is a pure Python implementation of multicast DNS service discovery. Prior to 0.149.16, _read_character_string and _read_string in src/zeroconf/_protocol/incoming.py advanced self.offset by attacker-declared RDLENGTH without checking it against self._data_len, allowing unauthenticated hosts on the local link over UDP/5353 (224.0.0.251 / ff02::fb) to send a TXT, HINFO, or A/AAAA record with rdlength=65535 and seed DNSCache and ServiceInfo.properties with truncated, attacker-shaped key/value or address records. This issue is fixed in version 0.149.16.
Zeroconf is a pure Python implementation of multicast DNS service discovery. Prior to 0.149.12, AsyncListener.handle_query_or_defer retained every truncated TC-bit incoming query, each up to _MAX_MSG_ABSOLUTE = 8966 bytes, in self._deferred[addr] and armed a per-address timer in self._timers[addr] without capping the per-address list or distinct addr keys, allowing unauthenticated hosts on the local link over UDP/5353 (224.0.0.251 / ff02::fb) to spoof sources, grow _deferred and _timers, and cause memory exhaustion and quadratic CPU burn. This issue is fixed in version 0.149.12.
Zeroconf: Unbounded DNS record cache allows LAN-local memory exhaustion via multicast flood
Zeroconf is a pure Python implementation of multicast DNS service discovery. Prior to 0.149.7, DNSCache._async_add inserted every response record into cache, _expirations, _expire_heap, and service_cache without a cap, allowing unauthenticated hosts on the local link over UDP/5353 (224.0.0.251 / ff02::fb) to multicast valid mDNS responses with unique names and cause memory exhaustion, slower cache lookups, slower async_expire passes, and broken discovery, registration, and ServiceBrowser callbacks. This issue is fixed in version 0.149.7.
Zeroconf: Unbounded exception-dedup state retains packet buffers via traceback frame locals, enabling LAN-local memory exhaustion
Zeroconf is a pure Python implementation of multicast DNS service discovery. Prior to 0.149.6, DNSIncoming._log_exception_debug and the four QuietLogger exception-dedup methods stored an unbounded _seen_logs dictionary keyed by attacker-influenced IncomingDecodeError messages, retaining sys.exc_info() tracebacks whose frame locals kept raw packet self.data buffers and allowing unauthenticated hosts on the local link over UDP/5353 (224.0.0.251 / ff02::fb) to drive memory growth until mDNS-dependent features degrade or the process is OOM-killed. This issue is fixed in version 0.149.6.
Zeroconf: Unbounded recursion in DNS compression-pointer decoder allows LAN-local denial of service
Zeroconf is a pure Python implementation of multicast DNS service discovery. Prior to 0.149.5, DNSIncoming._decode_labels_at_offset recurses once per DNS-name compression pointer, and a single mDNS packet carrying chained pointers can trigger a RecursionError that escapes DNSIncoming.__init__, causing sustained CPU burn, log flooding, and degraded mDNS-dependent features for unauthenticated hosts on the local link over UDP/5353 (224.0.0.251 / ff02::fb). This issue is fixed in version 0.149.5.
pyzipper: Encryption bypass for small files encrypted with pyzipper
pyzipper is a replacement for Python's zipfile that can read and write AES encrypted zip files. Prior to 0.4.0, a Python operator precedence bug in pyzipper/zipfile_aes.py caused the AE-2 format to never be automatically selected during encryption, causing encrypted entries to be written in AE-1 format and exposing the plaintext CRC32 checksum in the ZIP header and, for unseekable zip archives, in the datadescripter section, allowing an attacker who possesses the archive to brute-force candidate plaintexts for small or low-entropy files by comparing CRC32 values. This issue is fixed in version 0.4.0.
wger IDOR: Authenticated Users Can Read Others' Private Workout Session Data via Template Routine API
wger is a free, open-source workout and fitness manager. In versions prior to 2.6, any authenticated user can read another user's private workout session notes, exercise history, and training statistics by calling the /logs/ and /stats/ actions on a routine they do not own. The vulnerability exists in RoutineViewSet (wger/manager/api/views.py). The view defines two custom actions /logs/ and /stats/ that are intended to return data for the requesting user's own training history within a routine. However, the underlying permission check (RoutinePermission.has_object_permission) grants read access to any authenticated user when the routine has i
wger: Privilege escalation via trainer-login session chaining allows gym trainers to impersonate gym managers
wger is a free, open-source workout and fitness manager. In versions prior to 2.6, a gym trainer can escalate their session to any higher-privileged account (gym manager, general manager) by chaining two calls to the trainer-login endpoint. Once a trainer performs a legitimate switch into a low-privileged user, the session flag trainer.identity is set and this flag alone bypasses the permission check on all subsequent trainer-login calls. This grants full gym administration capabilities including viewing all member data, modifying contracts, managing gym configuration, and accessing other trainers' and managers' personal information. This iss