FastMCP is the standard framework for building MCP applications. Prior to version 2.14.2, the server does not properly respect the resource parameter submitted by the client in the authorization and token request. Instead of issuing the token explicitly for the MCP server, the token is issued for the base_url passed to the OAuthProxy during initialization. This issue has been patched 2.14.2.

cryptography is a package designed to expose cryptographic primitives and recipes to Python developers. Prior to 46.0.5, the public_key_from_numbers (or EllipticCurvePublicNumbers.public_key()), EllipticCurvePublicNumbers.public_key(), load_der_public_key() and load_pem_public_key() functions do not verify that the point belongs to the expected prime-order subgroup of the curve. This missing validation allows an attacker to provide a public key point P from a small-order subgroup. This can lead to security issues in various situations, such as the most commonly used signature verification (ECDSA) and shared key negotiation (ECDH). When the victim computes the shared secret as S = [victim_private_key]P via ECDH, this leaks information about victim_private_key mod (small_subgroup_order). For curves with cofactor > 1, this reveals the least significant bits of the private key. When these weak public keys are used in ECDSA , it's easy to forge signatures on the small subgroup. Only SECT curves are impacted by this. This vulnerability is fixed in 46.0.5.

A flaw was found in libsoup. After a CONNECT tunnel is established through an HTTP proxy, libsoup incorrectly attaches the Proxy-Authorization header to subsequent HTTPS requests sent through that tunnel to the destination server. This allows the destination server to capture proxy credentials, leading to information disclosure.

Arbitrary file read in the model loading mechanism (HDF5 integration) in Keras versions 3.0.0 through 3.13.1 on all supported platforms allows a remote attacker to read local files and disclose sensitive information via a crafted .keras model file utilizing HDF5 external dataset references.

A flaw was found in Spacewalk, as used in Red Hat Network Satellite. This open redirect vulnerability allows remote attackers to redirect users to arbitrary web sites by manipulating a URL in the url_bounce parameter. This can enable attackers to conduct phishing attacks, potentially leading to unauthorized information disclosure or credential theft.

Information disclosure due to uninitialized memory in Firefox and Firefox Focus for Android. This vulnerability was fixed in Firefox 148.

A vulnerability was found in libX11 due to a boundary condition within the _XkbReadKeySyms() function. This flaw allows a local user to trigger an out-of-bounds read error and read the contents of memory on the system.

A flaw was found in Red Hat Quay and mirror registry for Red Hat OpenShift. The log export feature in these products allows an authenticated user to specify an arbitrary callback URL. A backend process then makes server-side HTTP requests to this provided URL. This vulnerability, known as Server-Side Request Forgery (SSRF), could allow an attacker to send requests from the application's internal network, potentially leading to the disclosure of sensitive information.

A flaw was found in Red Hat Quay's Proxy Cache configuration feature. When an organization administrator configures an upstream registry for proxy caching, Quay makes a network connection to the specified registry hostname without verifying that it points to a legitimate external service. An attacker with organization administrator privileges could supply a crafted hostname to force the Quay server to make requests to internal network services, cloud infrastructure endpoints, or other resources that should not be accessible from the Quay application.

Axios is a promise based HTTP client for the browser and Node.js. Versions prior to 1.15.0 and 0.3.1 are vulnerable to a specific gadget-style attack chain in which prototype pollution in a third-party dependency may be leveraged to inject unsanitized header values into outbound requests. This vulnerability is fixed in 1.15.0 and 0.3.1.

Axios is a promise based HTTP client for the browser and Node.js. Prior to 0.32.0 and 1.16.0, Axios does not normalise IPv4-mapped IPv6 addresses. When NO_PROXY lists an IPv4 address such as 127.0.0.1 or 169.254.169.254, a request URL using the IPv4-mapped IPv6 form (::ffff:7f00:1, ::ffff:a9fe:a9fe) still routes through the configured proxy. Node.js resolves these addresses to the underlying IPv4 host, so the request reaches the internal service via the proxy rather than being blocked. This vulnerability is fixed in 0.32.0 and 1.16.0.

WeasyPrint helps web developers to create PDF documents. Prior to version 68.0, a server-side request forgery (SSRF) protection bypass exists in WeasyPrint's `default_url_fetcher`. The vulnerability allows attackers to access internal network resources (such as `localhost` services or cloud metadata endpoints) even when a developer has implemented a custom `url_fetcher` to block such access. This occurs because the underlying `urllib` library follows HTTP redirects automatically without re-validating the new destination against the developer's security policy. Version 68.0 contains a patch for the issue.

A flaw was found in Keycloak. An authenticated attacker can perform Server-Side Request Forgery (SSRF) by manipulating the `client_session_host` parameter during refresh token requests. This occurs when a Keycloak client is configured to use the `backchannel.logout.url` with the `application.session.host` placeholder. Successful exploitation allows the attacker to make HTTP requests from the Keycloak server’s network context, potentially probing internal networks or internal APIs, leading to information disclosure.

Next.js is a React framework for building full-stack web applications. From 13.4.13 to before 15.5.16 and 16.2.5, self-hosted applications using the built-in Node.js server can be vulnerable to server-side request forgery through crafted WebSocket upgrade requests. An attacker can cause the server to proxy requests to arbitrary internal or external destinations, which may expose internal services or cloud metadata endpoints. Vercel-hosted deployments are not affected. This vulnerability is fixed in 15.5.16 and 16.2.5.

Docling simplifies document processing by parsing diverse formats and providing integrations with the generative AI ecosystem. FIn versions >= 2.82.0, < 2.91.0, if the HTML backend was explicitly configured for rendering (rendering option by default deactivated), then the Playwright-based rendering feature could allow JavaScript execution and unrestricted network access when processing untrusted HTML documents. An attacker could craft malicious HTML that executes arbitrary JavaScript in the rendering context or makes unauthorized network requests to internal services, potentially leading to SSRF attacks, data exfiltration, or remote code execution in the rendering environment. This vulnerability is fixed in 2.91.0.

A flaw was identified in Keycloak, an identity and access management solution, where it improperly follows HTTP redirects when processing certain client configuration requests. This behavior allows an attacker to trick the server into making unintended requests to internal or restricted resources. As a result, sensitive internal services such as cloud metadata endpoints could be accessed. This issue may lead to information disclosure and enable attackers to map internal network infrastructure.

OpenStack Glance before 29.1.1, 30.x before 30.1.1, and 31.0.0 is affected by Server-Side Request Forgery (SSRF). By use of HTTP redirects, an authenticated user can bypass URL validation checks and redirect to internal services. Only glance image import functionality is affected. In particular, the web-download and glance-download import methods are subject to this vulnerability, as is the optional (not enabled by default) ovf_process image import plugin.

FastMCP is a Pythonic way to build MCP servers and clients. Prior to version 3.2.0, the OpenAPIProvider in FastMCP exposes internal APIs to MCP clients by parsing OpenAPI specifications. The RequestDirector class is responsible for constructing HTTP requests to the backend service. A vulnerability exists in the _build_url() method. When an OpenAPI operation defines path parameters (e.g., /api/v1/users/{user_id}), the system directly substitutes parameter values into the URL template string without URL-encoding. Subsequently, urllib.parse.urljoin() resolves the final URL. Since urljoin() interprets ../ sequences as directory traversal, an attacker controlling a path parameter can perform path traversal attacks to escape the intended API prefix and access arbitrary backend endpoints. This results in authenticated SSRF, as requests are sent with the authorization headers configured in the MCP provider. This issue has been patched in version 3.2.0.

A flaw was found in the FTP GVfs backend. A malicious FTP server can exploit this vulnerability by providing an arbitrary IP address and port in its passive mode (PASV) response. The client unconditionally trusts this information and attempts to connect to the specified endpoint, allowing the malicious server to probe for open ports accessible from the client's network.

vLLM is an inference and serving engine for large language models (LLMs). The SSRF protection fix for CVE-2026-24779 add in 0.15.1 can be bypassed in the load_from_url_async method due to inconsistent URL parsing behavior between the validation layer and the actual HTTP client. The SSRF fix uses urllib3.util.parse_url() to validate and extract the hostname from user-provided URLs. However, load_from_url_async uses aiohttp for making the actual HTTP requests, and aiohttp internally uses the yarl library for URL parsing. This vulnerability in 0.17.0.

Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 0.0.26 to before 1.56.0, aServer-Side Request Forgery (SSRF) vulnerability exists in Pydantic AI's URL download functionality. When applications accept message history from untrusted sources, attackers can include malicious URLs that cause the server to make HTTP requests to internal network resources, potentially accessing internal services or cloud credentials. This vulnerability only affects applications that accept message history from external users. This vulnerability is fixed in 1.56.0.

LangSmith Client SDKs provide SDK's for interacting with the LangSmith platform. The LangSmith SDK's distributed tracing feature is vulnerable to Server-Side Request Forgery via malicious HTTP headers. An attacker can inject arbitrary api_url values through the baggage header, causing the SDK to exfiltrate sensitive trace data to attacker-controlled endpoints. When using distributed tracing, the SDK parses incoming HTTP headers via RunTree.from_headers() in Python or RunTree.fromHeaders() in Typescript. The baggage header can contain replica configurations including api_url and api_key fields. Prior to the fix, these attacker-controlled values were accepted without validation. When a traced operation completes, the SDK's post() and patch() methods send run data to all configured replica URLs, including any injected by an attacker. This vulnerability is fixed in version 0.6.3 of the Python SDK and 0.4.6 of the JavaScript SDK.

vLLM is an inference and serving engine for large language models (LLMs). Prior to version 0.14.1, a Server-Side Request Forgery (SSRF) vulnerability exists in the `MediaConnector` class within the vLLM project's multimodal feature set. The load_from_url and load_from_url_async methods obtain and process media from URLs provided by users, using different Python parsing libraries when restricting the target host. These two parsing libraries have different interpretations of backslashes, which allows the host name restriction to be bypassed. This allows an attacker to coerce the vLLM server into making arbitrary requests to internal network resources. This vulnerability is particularly critical in containerized environments like `llm-d`, where a compromised vLLM pod could be used to scan the internal network, interact with other pods, and potentially cause denial of service or access sensitive data. For example, an attacker could make the vLLM pod send malicious requests to an internal `llm-d` management endpoint, leading to system instability by falsely reporting metrics like the KV cache state. Version 0.14.1 contains a patch for the issue.

A flaw was found in Red Hat Quay's notification webhook feature. The Slack and generic webhook notification handlers accept user-supplied URLs without SSRF validation, allowing a repository administrator to make the Quay worker issue POST requests to internal network addresses or cloud infrastructure endpoints that should not be reachable from the application.

A flaw was found in the `guardrails-detectors` component. This vulnerability allows a remote attacker to perform a blind Server-Side Request Forgery (SSRF) by submitting a specially crafted XML Schema Definition (XSD) string. This can lead to unauthorized access to sensitive information, including credentials from cloud metadata services, Kubernetes API, internal MinIO, and other internal network endpoints. Additionally, it enables local file reads of critical data such as service account tokens and pod secrets.

A flaw was found in Red Hat Quay's repository-level mirror configuration feature. The POST and PUT handlers in endpoints/api/mirror.py accept an external_reference parameter without SSRF validation, unlike the organization-level mirror handlers which apply validate_external_registry_url(). A repository administrator can supply a crafted hostname that causes the Quay mirror worker to make requests via Skopeo to internal network services, cloud metadata endpoints, or other resources not intended to be reachable from the Quay application.

A server-side request forgery (SSRF) flaw was found in KubeVirt's virt-api port-forward handler. When processing a port-forward request to a VirtualMachineInstance (VMI), virt-api reads the target IP from vmi.Status.Interfaces[0].IP and passes it directly to net.Dial() without validation. For VMIs using non-masquerade network bindings (bridge or secondary-only), this IP is reported by the QEMU guest agent running inside the VM and is fully controllable by the VM owner. An attacker with kubevirt.io:edit permissions can create a VM with a modified guest agent that reports an arbitrary IP address, then request port-forward to establish a bidirectional TCP tunnel from virt-api's cluster-internal network position to any routable destination, bypassing NetworkPolicy isolation.

A flaw was identified in Keycloak’s OpenID Connect Dynamic Client Registration feature when clients authenticate using private_key_jwt. The issue allows a client to specify an arbitrary jwks_uri, which Keycloak then retrieves without validating the destination. This enables attackers to coerce the Keycloak server into making HTTP requests to internal or restricted network resources. As a result, attackers can probe internal services and cloud metadata endpoints, creating an information disclosure and reconnaissance risk.

A flaw was found in the AWX GitHub webhook integration. When processing GitHub pull_request webhooks, the controller stores the pull_request.statuses_url value from the webhook payload without validating that it points to a trusted GitHub API endpoint. If a job template is configured with a GitHub Personal Access Token as its webhook credential, the controller later POSTs that token to the stored callback URL when posting job status updates. An attacker who can submit a correctly signed forged webhook using the job template's webhook_key can redirect the callback to an attacker-controlled URL and exfiltrate the configured GitHub PAT.

A flaw was found in Apicurio Registry. The WSDLReaderAccessor creates a wsdl4j WSDLReader without disabling the javax.wsdl.importDocuments feature. When the VALIDITY rule is set to FULL, an attacker with Developer-role access can upload a WSDL document containing attacker-controlled import locations, causing the registry to issue HTTP requests to arbitrary internal URLs (server-side request forgery).

LangChain before 0.0.317 allows SSRF via document_loaders/recursive_url_loader.py because crawling can proceed from an external server to an internal server.

A flaw was found in the Quay config-tool's LDAP and SMTP validation functions. An attacker with config editor access can exploit these functions, which make outbound connections to user-supplied endpoints without proper IP or host filtering. This allows the attacker to perform internal network reconnaissance from the Quay pod's network position, potentially mapping the internal network infrastructure.

External Control of File Name or Path (CWE-73) combined with Server-Side Request Forgery (CWE-918) can allow an attacker to cause arbitrary file disclosure through a specially crafted credentials JSON payload in the Google Gemini connector configuration. This requires an attacker to have authenticated access with privileges sufficient to create or modify connectors (Alerts & Connectors: All). The server processes a configuration without proper validation, allowing for arbitrary network requests and for arbitrary file reads.

LangChain is a framework for building LLM-powered applications. Prior to version 1.1.8, a redirect-based Server-Side Request Forgery (SSRF) bypass exists in `RecursiveUrlLoader` in `@langchain/community`. The loader validates the initial URL but allows the underlying fetch to follow redirects automatically, which permits a transition from a safe public URL to an internal or metadata endpoint without revalidation. This is a bypass of the SSRF protections introduced in 1.1.14 (CVE-2026-26019). Users should upgrade to `@langchain/community` 1.1.18, which validates every redirect hop by disabling automatic redirects and re-validating `Location` targets before following them. In this version, automatic redirects are disabled (`redirect: "manual"`), each 3xx `Location` is resolved and validated with `validateSafeUrl()` before the next request, and a maximum redirect limit prevents infinite loops.

If kdcproxy receives a request for a realm which does not have server addresses defined in its configuration, by default, it will query SRV records in the DNS zone matching the requested realm name. This creates a server-side request forgery vulnerability, since an attacker could send a request for a realm matching a DNS zone where they created SRV records pointing to arbitrary ports and hostnames (which may resolve to loopback or internal IP addresses). This vulnerability can be exploited to probe internal network topology and firewall rules, perform port scanning, and exfiltrate data. Deployments where the "use_dns" setting is explicitly set to false are not affected.

A Server-Side Request Forgery (SSRF) vulnerability exists in the RequestsToolkit component of the langchain-community package (specifically, langchain_community.agent_toolkits.openapi.toolkit.RequestsToolkit) in langchain-ai/langchain version 0.0.27. This vulnerability occurs because the toolkit does not enforce restrictions on requests to remote internet addresses, allowing it to also access local addresses. As a result, an attacker could exploit this flaw to perform port scans, access local services, retrieve instance metadata from cloud environments (e.g., Azure, AWS), and interact with servers on the local network. This issue has been fixed in version 0.0.28.

A security flaw has been discovered in vibrantlabsai RAGAS up to 0.4.3. The affected element is the function _try_process_local_file/_try_process_url of the file src/ragas/metrics/collections/multi_modal_faithfulness/util.py of the component Collections Module. Performing a manipulation of the argument retrieved_contexts results in server-side request forgery. The attack can be initiated remotely. The exploit has been released to the public and may be used for attacks. The security patch for CVE-2025-45691 was applied to a different module only. The vendor was contacted early about this disclosure but did not respond in any way.

An Arbitrary File Read vulnerability exists in the ImageTextPromptValue class in Exploding Gradients RAGAS v0.2.3 to v0.2.14. The vulnerability stems from improper validation and sanitization of URLs supplied in the retrieved_contexts parameter when handling multimodal inputs.

A flaw was found in OpenShift Console. A Server Side Request Forgery (SSRF) attack can happen if an attacker supplies all or part of a URL to the server to query. The server is considered to be in a privileged network position and can often reach exposed services that aren't readily available to clients due to network filtering. Leveraging such an attack vector, the attacker can have an impact on other services and potentially disclose information or have other nefarious effects on the system. The /api/dev-console/proxy/internet endpoint on the OpenShift Console allows authenticated users to have the console's pod perform arbitrary and fully controlled HTTP(s) requests. The full response to these requests is returned by the endpoint. While the name of this endpoint suggests the requests are only bound to the internet, no such checks are in place. An authenticated user can therefore ask the console to perform arbitrary HTTP requests from outside the cluster to a service inside the cluster.

With the following crawler configuration: ```python from bs4 import BeautifulSoup as Soup url = "https://example.com" loader = RecursiveUrlLoader( url=url, max_depth=2, extractor=lambda x: Soup(x, "html.parser").text ) docs = loader.load() ``` An attacker in control of the contents of `https://example.com` could place a malicious HTML file in there with links like "https://example.completely.different/my_file.html" and the crawler would proceed to download that file as well even though `prevent_outside=True`. https://github.com/langchain-ai/langchain/blob/bf0b3cc0b5ade1fb95a5b1b6fa260e99064c2e22/libs/community/langchain_community/document_loaders/recursive_url_loader.py#L51-L51 Resolved in https://github.com/langchain-ai/langchain/pull/15559

A flaw was found in ose-openshift-apiserver. This vulnerability allows internal network enumeration, service discovery, limited information disclosure, and potential denial-of-service (DoS) through Server-Side Request Forgery (SSRF) due to missing IP address and network-range validation when processing user-supplied image references.

Starlette is a lightweight ASGI framework/toolkit. In versions 1.0.1 and earlier, StaticFiles on Windows is vulnerable to SSRF. An UNC path such as \\attacker.com\share can cause os.path.realpath to initiate an outbound SMB connection before the path is rejected, exposing the service account’s NTLMv2 credentials for offline cracking or relay even though the HTTP response is only a 404. The issue affects default follow_symlink=False deployments, including frameworks built on Starlette such as FastAPI; POSIX systems and follow_symlink=True are unaffected. The issue is fixed in 1.1.0.

Angular is a development platform for building mobile and desktop web applications using TypeScript/JavaScript and other languages. Prior to 22.0.0-next.12, 21.2.13, 20.3.21, and 19.2.22, a Server-Side Request Forgery (SSRF) vulnerability exists in @angular/platform-server. The issue stems from how the server-side rendering (SSR) engine processes the request URL provided to the rendering entry points. When an absolute-form URL (e.g., http://evil.com) is passed to the rendering engine, the internal ServerPlatformLocation can be manipulated into adopting the attacker-controlled domain as the "current" hostname. Consequently, any relative HttpClient requests or PlatformLocation.hostname references are redirected to the attacker controlled server, potentially exposing internal APIs or metadata services. This vulnerability is fixed in 22.0.0-next.12, 21.2.13, 20.3.21, and 19.2.22.

A flaw was found in the OpenShift Router. A user with EndpointSlice write access can exploit this vulnerability by creating a Service backed by an FQDN (Fully Qualified Domain Name) EndpointSlice that resolves to a cloud metadata endpoint. This allows the router to proxy requests to the cloud metadata endpoint, leading to the disclosure of instance credentials and other sensitive metadata. This bypasses previous security measures for validating IP addresses.

Axios is a promise based HTTP client for the browser and Node.js. Prior to 1.15.1 and 0.31.1, an attacker who can influence the target URL of an Axios request can use any address in the 127.0.0.0/8 range (other than 127.0.0.1) to completely bypass the NO_PROXY protection. This vulnerability is due to an incomplete for CVE-2025-62718, This vulnerability is fixed in 1.15.1 and 0.31.1.

LangChain is a framework for building agents and LLM-powered applications. Prior to 1.1.14, langchain-openai's _url_to_size() helper (used by get_num_tokens_from_messages for image token counting) validated URLs for SSRF protection and then fetched them in a separate network operation with independent DNS resolution. This left a TOCTOU / DNS rebinding window: an attacker-controlled hostname could resolve to a public IP during validation and then to a private/localhost IP during the actual fetch.