Pip vulnerabilities

CVE-2026-55787 High · 7.1 CWE-918 Oct 14, 2026

Flyto2 Core: SSRF guard bypass via IPv6 transition addresses (IPv4-mapped / 6to4 / NAT64) in validate_url_ssrf

Flyto2 Core is an execution kernel for automation and AI-agent workflows. Prior to 2.26.3, validate_url_ssrf and is_private_ip in src/core/utils.py compare resolved addresses only against native private IPv4 and IPv6 ranges and do not unwrap IPv4-mapped, IPv4-compatible, 6to4, or NAT64 transition addresses. An authenticated workflow author can provide a transition-form literal URL through the Execution API POST /v1/execute route, http.get, or sibling guarded outbound-request modules, causing the guard to accept an address whose embedded IPv4 points to loopback, RFC 1918, or cloud metadata services. On a host that routes the selected transition form, the outbound request reaches the internal destination and returns the response body to the workflow caller, enabling read SSRF and disclosure of internal-service data or cloud credentials. This issue is fixed in version 2.26.3.

CVE-2026-55786 High · 8.4 CWE-78 Oct 14, 2026

Flyto2 Core: Unauthenticated Command Execution via HTTP MCP `execute_module`

Flyto2 Core is an execution kernel for automation and AI-agent workflows. From 2.26.2 until 2.26.4, the HTTP MCP POST /mcp route accepts unauthenticated JSON-RPC tools/call requests and forwards execute_module arguments to the module registry without the require_auth dependency used by the REST execution routes. An unauthenticated caller can select sandbox.execute_shell and supply its command parameter, which reaches asyncio.create_subprocess_shell with the privileges of the flyto-core server process. The endpoint is reachable locally under the default loopback binding and becomes network reachable when an operator binds the server to a non-loopback interface, allowing arbitrary OS command execution and potential full host compromise. This issue is fixed in version 2.26.4.

CVE-2026-57118 Critical · 9.8 CWE-306 Oct 12, 2026

PraisonAI AgentTeam.launch exposes unauthenticated remote agent listing and invocation endpoints

PraisonAI is a multi-agent teams system. Prior to praisonaiagents 1.6.59, AgentTeam.launch and the Agents alias register GET /{path}/list, POST /{path}, and `POST /{path}/{agent_id}' without access-control checks before listing agents or calling agent.chat. A remotely exposed server can therefore let unauthenticated clients enumerate agents, consume model resources, access connected private context, and trigger agent tools or workflow side effects. This vulnerability is fixed in praisonaiagents 1.6.59.

CVE-2026-48484 Medium · 6.5 CWE-20 Oct 9, 2026

pyLoad: Lack of Input Size Validation Leads to Denial of Service (DoS) and Process Termination

pyLoad is a free and open-source download manager written in Python. Prior to 0.5.0b3.dev101, the API `rpc` function in `api_blueprint.py` handles `multipart/form-data` uploads by reading the whole content of the uploaded file into memory with `file.read()`. This occurs before the data is sent to the underlying function. Since there is no size limit set at this point, a large file upload can exhaust the server's available memory which led to process termination. Version 0.5.0b3.dev101 contains a patch.

CVE-2026-107728 High · 7.5 CWE-863 Oct 8, 2026

Strawberry GraphQL: Synchronous permission checks can treat an awaitable authorization result as truthy

Strawberry GraphQL is a library for creating GraphQL APIs. From 0.217.0 until 0.326.1, PermissionExtension.resolve() on a synchronous field resolver evaluates the result of has_permission() for truthiness. When a custom permission declares has_permission() as a normal function but returns an awaitable, supports_sync does not classify it as asynchronous, the awaitable is not awaited, and its inherently truthy object value permits the protected resolver to run even when the result would resolve to false. This affects synchronous field resolvers under both execute_sync() and execute(); permissions declared with async def has_permission() and synchronous permissions returning a boolean are not affected. This issue is fixed in version 0.326.1.

CVE-2026-107377 High · 7.5 CWE-22 Oct 8, 2026

datamodel-code-generator: Protobuf weak-import path traversal allows files to be written outside the temporary directory

datamodel-code-generator generates Python data models from schema definitions. From 0.59.0 until 0.81.0, an attacker-controlled Protobuf schema can supply absolute or parent-directory paths captured by WEAK_IMPORT_PATTERN and consumed by _write_missing_weak_imports in src/datamodel_code_generator/parser/protobuf.py. Exploitation requires a victim or automated job to process the attacker-controlled schema with Protobuf input support, which requires the grpcio-tools package. The paths escape the weak_imports temporary directory before protoc runs, allowing creation of directory trees and new files or overwrite of existing writable files with a generated Protobuf syntax declaration. The effect persists when later Protobuf compilation fails. The written content is limited to a proto2 or proto3 syntax declaration, and direct arbitrary code execution has not been demonstrated. This issue is fixed in version 0.81.0.

CVE-2026-107292 Medium · 6.4 CWE-346 Oct 8, 2026

Pydantic AI Web chat UI (`Agent.to_web()`, `clai web`): the local chat endpoint does not validate the `Host` header

Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.34.0 until 2.30.0, the Agent.to_web() and clai web development chat server does not validate the Host header, allowing a website visited by a developer to use DNS rebinding to reach a loopback-hosted agent as a same-origin service. The hostile page can read the served UI and submit chat requests that execute agent tools with the local process's privileges and credentials, causing data disclosure or unwanted side effects. Binding to localhost, Origin checks, and CSRF tokens do not prevent the same-origin DNS rebinding path. This issue is fixed in versions 1.107.5 and 2.30.0.

CVE-2026-107287 Medium · 6.5 CWE-400 Oct 8, 2026

Pydantic AI: Excessive resource use when local web fetching converts nested HTML

Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.77.0 until 1.107.7 and 2.52.0, the local web_fetch_tool and the WebFetch local fallback can consume excessive CPU and memory during HTML-to-Markdown conversion of attacker-controlled HTML containing deeply nested block elements. Conversion repeatedly reprocesses accumulated text and can greatly expand intermediate output before the returned-content limit is applied, allowing a model-directed fetch to delay other work in the process. Provider-native web fetching is not affected. This issue is fixed in versions 1.107.7 and 2.52.0.

CVE-2026-97688 Medium · 5.3 CWE-835 Sep 29, 2026

urllib3: Chunked Deflate streaming can enter an infinite loop

urllib3 is an HTTP client library for Python. From 2.6.2 until 2.8.0, HTTPResponse.stream and HTTPResponse.read_chunked can enter an infinite loop because the Deflate decoder retains trailing bytes as unconsumed input after reaching end-of-stream and repeatedly decodes them without progress. The issue occurs when an untrusted server sends a chunked Deflate response whose decoded body exceeds a positive finite chunk size and whose encoded body has trailing bytes, specifically a response with Transfer-Encoding: chunked and Content-Encoding: deflate, content decoding enabled, and the positive finite amt=N streaming chunk size. The attack mechanism is that a malicious server returns a compressed chunked response with trailing bytes after the Deflate stream. The impact is excessive CPU usage and a request that does not complete, and network read timeouts do not interrupt the loop because no further socket read occurs. This issue is fixed in version 2.8.0.

CVE-2026-97687 High · 8.1 CWE-295 Sep 29, 2026

urllib3: HTTPS proxy TLS configuration may be ignored or overridden

urllib3 is an HTTP client library for Python. From 1.26.0 until 2.8.0, the proxy_ssl_context, proxy_assert_hostname, proxy_assert_fingerprint, ssl_context, cert_reqs, verify_mode, use_forwarding_for_https=True, and CERT_NONE configuration paths fail to remain separated because target-server TLS settings are incorrectly applied to the HTTPS proxy connection. The trigger is that an application uses an HTTPS proxy and configures target-server TLS settings that must remain separate from the proxy TLS handshake, including HTTPS forwarding with target-specific identity or credentials. Applying cert_reqs=CERT_NONE can overwrite proxy_ssl_context.verify_mode in place, and the mutation persists so later connections reusing the same context may connect to the HTTPS proxy without certificate verification. The attack mechanism is that an attacker intercepts and impersonates the HTTPS proxy after the effective proxy policy accepts the attacker's certificate. The impact is that the attacker can observe or modify forwarded traffic or receive a target TLS client certificate, while CONNECT tunneling still preserves the separate end-to-end target TLS connection. This issue is fixed in version 2.8.0.

CVE-2026-97689 High · 7.5 CWE-770 Sep 29, 2026

urllib3: HTTPResponse.stream()/read_chunked() buffers an unbounded chunk-size line into memory

urllib3 is an HTTP client library for Python. From 1.10.3 until 2.8.0, the HTTPResponse.read_chunked and HTTPResponse.stream methods can allocate unbounded memory because the streaming chunk parser buffers the chunk-size field until newline or EOF without a length bound. The trigger is that a malicious server returns Transfer-Encoding: chunked followed by a very long run of bytes without a newline. The attack mechanism is that a malicious HTTP server sends a very long unterminated chunk-size line. The impact is that unbounded memory allocation can exhaust the client process. This issue is fixed in version 2.8.0.

CVE-2025-66455 Critical · 9.8 CWE-502 Sep 18, 2026

LMDeploy has Remote Code Execution by Pickle Deserialization via handle_zmq_recv in lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.py

LMDeploy is a toolkit for compressing, deploying, and serving large language models. Starting in version 0.9.2 and prior to version 0.16.0, LMDeploy's PyTorch DistServe/PD-disaggregation control plane used `recv_pyobj()` to deserialize messages received through a ZeroMQ PULL socket. PyZMQ implements `recv_pyobj()` using Python pickle deserialization, which can execute arbitrary code while reconstructing an object. The peer address used by the receiver was supplied through the `POST /distserve/p2p_connect` HTTP endpoint. An attacker who could reach an affected DistServe API server could cause the server to connect to an attacker-controlled ZeroMQ endpoint and deserialize a crafted pickle payload. API-key authentication is not enabled unless the operator explicitly configures it. As a result, affected DistServe deployments without API keys allowed unauthenticated remote code execution with the privileges of the LMDeploy serving process. This issue affects the PyTorch backend when PD-disaggregation/DistServe is enabled. Ordinary deployments that do not use the affected disaggregated-serving path do not expose this data flow. The fix was released in LMDeploy 0.16.0. Users who cannot upgrade immediately should prevent untrusted clients from reaching `/distserve/*` endpoints, restrict the DistServe HTTP and ZeroMQ control planes to trusted cluster networks, configure API-key authentication, and block arbitrary outbound ZeroMQ connections from serving nodes. These measures reduce exposure but do not make pickle deserialization safe.

CVE-2025-59953 Critical · 9.8 CWE-502 Sep 16, 2026

LMdeploy has Remote Code Execution by Pickle Deserialization via zmq_rpc.call_and_response() in InterLM/lmdeploy

LMDeploy is a toolkit for compressing, deploying, and serving large language models. Starting in version 0.9.1 and prior to version 0.10.2, the LMdeploy implements an rpc server (AsyncRPCServer in zmq_rpc.py) for supporting the RPC communications. In its core functionality call_and_response(), I found it will directly use the pickles.loads() to deserialize the received messages without any sanitization, hence resulting in a remote code execution vulnerability by this RPC server. Version 0.10.2 contains a patch.

CVE-2026-79657 Critical · 9.8 CWE-502 Aug 25, 2026

NLTK before 3.10.3 Remote Code Execution via Unsafe Pickle Deserialization

NLTK versions before 3.10.3 contain a remote code execution vulnerability in allowlisted pickle loaders that trust entire module namespaces instead of specific safe callables. Attackers can craft malicious pickle payloads invoking dangerous in-namespace functions like ReppTokenizer._execute and numpy.f2py.crackfortran.myeval through pickle REDUCE to execute arbitrary commands during model or tokenizer artifact loading.

CVE-2026-78682 High · 7.5 CWE-918 Aug 24, 2026

NLTK before 3.10.3 SSRF Protection Bypass via Proxy

NLTK before 3.10.3 contains a server-side request forgery vulnerability in nltk.pathsec.urlopen (and callers nltk.data.load, nltk.downloader.Downloader.index/download) when an HTTP proxy is configured. pathsec.urlopen validates the requested hostname locally, but proxy-handler inheritance disables the safe HTTP/HTTPS handlers so the actual fetch is performed by the proxy against a destination that is never re-validated. An attacker can supply a validated public URL that the proxy forwards to an internal loopback-only service, allowing disclosure of internal HTTP resources, loading of forged downloader indexes, and installation of attacker-chosen package content.

CVE-2026-70626 Medium · 6.2 CWE-59 Aug 22, 2026

NLTK before 3.9.4 Symlink Escape via CorpusReader

NLTK versions before 3.9.4 contain a symlink escape vulnerability in CorpusReader.open() that allows local attackers to read arbitrary files outside the corpus root. The vulnerability exists because path validation is lexical and does not account for symlink resolution, enabling attackers to place symlinks inside the corpus root to access files outside the intended boundary.

CVE-2026-66393 High · 7.5 CWE-674 Aug 22, 2026

NLTK before 3.9.4 Denial of Service via JSONTaggedDecoder

NLTK versions before 3.9.4 contain an unbounded recursion vulnerability in JSONTaggedDecoder.decode_obj() that allows attackers to cause denial of service by supplying deeply nested JSON structures. Attackers can craft JSON payloads exceeding the recursion limit to trigger an unhandled RecursionError that crashes the Python process.

CVE-2026-65915 Medium · 6.5 CWE-284 Aug 22, 2026

NLTK before 3.10.0 Arbitrary File Read via FileSystemPathPointer

NLTK versions before 3.10.0 contain a logic bug in FileSystemPathPointer.open() where the sandbox validation check compares a normalized path against itself, making the security check permanently inert. Attackers can pass file:// URLs to nltk.data.load() to read arbitrary files accessible to the process user, including credentials and configuration files.

CVE-2026-63312 High · 7.5 CWE-22 Aug 22, 2026

NLTK StreamBackedCorpusView Bypasses pathsec.ENFORCE Arbitrary File Read

NLTK before 3.10.0 contains an arbitrary local file read vulnerability in StreamBackedCorpusView that bypasses pathsec.ENFORCE by calling builtins.open() directly instead of pathsec.open(). Attackers who control the fileid argument can read arbitrary local files regardless of the ENFORCE setting, including sensitive system files and application credentials.

CVE-2026-63311 Medium · 5.3 CWE-918 Aug 22, 2026

NLTK before 3.10.0 SSRF via DNS Resolution Failure

NLTK before 3.10.0 (affected versions <= 3.9.4) contains a server-side request forgery (SSRF) vulnerability in the validate_network_url() function in nltk/pathsec.py. The _resolve_hostname() helper catches OSError and ValueError during socket.getaddrinfo() and returns an empty list; when DNS resolution fails, the validation loop executes no IP checks and the function fails open, allowing urlopen() to proceed without validation. An attacker who can trigger DNS resolution failures or use DNS rebinding can bypass SSRF protections and reach restricted network resources, including cloud metadata endpoints (e.g., 169.254.169.254).

CVE-2026-62388 High · 7.5 CWE-1188 Aug 22, 2026

NLTK before 3.10.0 Insecure Default Configuration in pathsec.py

NLTK versions before 3.10.0 default to ENFORCE=False in pathsec.py, causing all security validation functions to emit warnings instead of raising exceptions. Attackers can bypass path traversal and pickle deserialization protections by exploiting the disabled security controls that are only active when manually enabled.

CVE-2026-62385 High · 7.5 CWE-73 Aug 22, 2026

NLTK 3.9.4 Path Traversal via FrameNet and NKJP Readers

NLTK versions before 3.10.0 contain a path traversal vulnerability in FramenetCorpusReader and NKJPCorpusReader that allows attackers to parse XML files outside the corpus root by supplying unsafe selectors or poisoned index state. Attackers can exploit frame_by_name, doc, lu, and header methods with crafted parameters to read arbitrary XML files accessible to the application.

CVE-2026-62384 High · 7.5 CWE-22 Aug 22, 2026

NLTK FramenetCorpusReader Symlink Sandbox Bypass before 3.10.2

NLTK versions before 3.10.2 contain a symlink-based sandbox bypass in FramenetCorpusReader that allows attackers to read arbitrary XML files outside the corpus root. Attackers can place symlinks with names containing no path separators inside the corpus subdirectory, which pass the path validation guard and are resolved to files outside the intended corpus root when accessed via frame_by_name(), _lu_file(), or doc() methods.

CVE-2026-62383 Medium · 5.5 CWE-22 Aug 22, 2026

nltk IPIPANCorpusReader Symlink Arbitrary File Read

nltk versions before 3.10.2 contain a symlink-based arbitrary file read vulnerability in IPIPANCorpusReader methods that bypass nltk.pathsec validation entirely. Attackers can place a symlink in the corpus root directory and read arbitrary files accessible to the process by calling channels(), domains(), categories(), or fileids() methods with the symlink filename.

CVE-2026-71514 Low · 3.3 CWE-22 Aug 22, 2026

NLTK 3.9.4 through 3.10.2 Path Traversal via CrubadanCorpusReader pathsec Bypass

NLTK 3.9.4 through 3.10.2 contains a path traversal vulnerability in CrubadanCorpusReader. _load_lang_ngrams joins the corpus root with crubadan_code, the column-0 value read from the corpus table.txt mapping file, and opens the result with the builtin open() rather than the pathsec-validated opener, so os.path.join discards the root when that value is absolute and the read escapes the corpus directory without the containment check nltk.pathsec applies when ENFORCE is set. An attacker who controls a corpus package can disclose file contents outside the corpus root through lang_freq, limited to paths ending in -3grams.txt whose contents parse as token count lines.