| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 0.3.4 until 1.107.4 and 2.27.1, OpenTelemetry instrumentation configured with InstrumentationSettings(include_content=False) can export retry prompts outside tool calls in gen_ai.input.messages and pydantic_ai.all_messages. Agents using NativeOutput, PromptedOutput, or output validators on text output can therefore disclose validation feedback, including invalid model values quoted by that feedback, to readers of the telemetry backend. Tool-call retries and deployments that do not use include_content=False are not affected by this specific path. This issue is fixed in versions 1.107.4 and 2.27.1. |
| 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. |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 0.3.4 until 1.107.6 and 2.44.0, OpenTelemetry instrumentation configured with InstrumentationSettings(include_content=False) can still export sensitive agent content through exception.message and exception.stacktrace events, error status descriptions, and model_request_parameters containing instructions or the prompted_output_template. The exposed data is available to readers of the configured telemetry backend and can include tool feedback, provider error bodies, runtime instructions, and structured-output templates even though message attributes are redacted. This issue does not grant new access to agent data, and deployments that do not use include_content=False are not affected by the setting bypass. This issue is fixed in versions 1.107.6 and 2.44.0. |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.77.0 until 1.107.6 and 2.44.0, the local web_fetch_tool and the WebFetch local fallback compare blocked_domains entries with a URL hostname before both values are normalized to the form used by getaddrinfo. An attacker-influenced model can use an equivalent IDNA spelling, non-ASCII label separator, case variation, or trailing root label that resolves to a blocked host but does not match the configured string, causing the application to fetch that host with its own privileges. allowed_domains fails closed for unmatched spellings, and private-IP and cloud-metadata protections remain effective. This issue is fixed in versions 1.107.6 and 2.44.0. |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 2.10.0 until 2.53.0, streamed requests made through ConcurrencyLimitedModel or limit_model_concurrency can retain shared concurrency slots because anyio.CapacityLimiter associates an acquired slot with the borrowing task while streaming cleanup can run in a different task. Early stream termination, cancellation, consumer exceptions, or complete stream_text() consumption with debounce_by=0.1 can therefore leave capacity occupied, eventually preventing later requests that share the long-lived limiter from proceeding and causing a denial of service. Agent-level max_concurrency and non-streaming model requests are not affected. This issue is fixed in version 2.53.0. |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.56.0 until 1.107.6 and 2.44.0, applications that opt attacker-influenced URLs into local network access through FileUrl with force_download='allow-local' or web_fetch_tool with allow_local_urls=True can bypass the cloud-metadata blocklist by appending an IPv6 zone identifier to an IPv6 metadata address. IPv6Address equality and hashing include the zone identifier, so the blocklist comparison fails even though the network stack ignores the zone on a non-link-local destination and reaches the metadata service, potentially exposing cloud IAM credentials. The opt-in settings are disabled by default, and the issue requires an IPv6-enabled environment. This issue is fixed in versions 1.107.6 and 2.44.0. |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.77.0 until 1.107.2 and 2.24.0, web_fetch_tool, the WebFetch local fallback, and remote FileUrl media downloads buffer the complete HTTP response body before enforcing content-size controls. An attacker-influenced URL can stream an arbitrarily large response that exhausts process memory and crashes the worker; affected media types include ImageUrl, DocumentUrl, VideoUrl, and AudioUrl. SSRF protections remain effective, and the impact is limited to availability. This issue is fixed in versions 1.107.2 and 2.24.0. |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.77.0 until 1.107.6 and 2.44.0, the local web_fetch_tool and the WebFetch local fallback process server-controlled responses with quadratic title extraction, whitespace normalization, and ordered-list numbering. An attacker-controlled page of modest size can therefore block the event loop for an extended period, stalling other agent runs and requests, while unsupported codecs or excessive HTML or JSON nesting can abort an individual run. This issue is fixed in versions 1.107.6 and 2.44.0. |
| 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. |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 1.34.0 until 1.107.4 and 2.28.0, the Agent.to_web() and clai web development chat endpoint has missing request content-type validation. A website visited by a developer can submit a browser-compatible request to a loopback-hosted chat server, causing the served agent to run and execute tools with the privileges and credentials of the local process; client-relayed approval decisions also leave requires_approval=True tools exposed. Binding to localhost does not prevent a browser page from reaching the loopback address. This issue is fixed in versions 1.107.4 and 2.28.0. |
| HTTPX2 is a next generation HTTP client for Python. Prior to 2.10.0, httpcore2 fails to start TLS in src/httpcore2/httpcore2/_sync/socks_proxy.py and src/httpcore2/httpcore2/_async/socks_proxy.py when the remote origin uses wss through a SOCKS5 proxy because the TLS upgrade condition only recognizes https. HTTPX2 exposes the flaw through Client.websocket() and AsyncClient.websocket() from 2.6.0 through 2.9.1, so the opening handshake, query parameters, Authorization headers, cookies, and subsequent frames can cross the proxy path in plaintext without certificate verification. An attacker controlling or observing that path can read or modify traffic and impersonate the WebSocket server. This issue is fixed in httpcore2 2.10.0 and HTTPX2 2.10.0. |
| HTTPX2 is a next generation HTTP client for Python. From 2.5.0 until 2.10.0, the HTTPX2 Server-Sent Events parser in src/httpx2/httpx2/_sse.py repeatedly copies and rescans buffered text in _SSELineDecoder.decode() when an attacker-controlled or compromised SSE endpoint splits one unterminated line across many response chunks. The behavior affects httpx2.Client.sse() and httpx2.AsyncClient.sse(), and the total processing work grows quadratically with the line length, allowing a crafted stream to consume excessive CPU and block a synchronous worker or asynchronous event loop. This issue is fixed in version 2.10.0. |
| HTTPX2 is a next generation HTTP client for Python. Prior to 2.11.0, FileField.render_headers() in src/httpx2/httpx2/_multipart.py directly interpolates attacker-controlled content_type values and custom headers from the files= three-element (filename, content, content_type) tuple and the files= four-element (filename, content, content_type, headers) tuple into multipart/form-data part headers without validating header names or values. CR or LF characters can terminate a part header, inject additional part headers, or end the part header block early, allowing a downstream multipart parser to treat attacker-supplied lines as genuine headers and potentially alter part semantics or bypass header-based checks. This issue is fixed in version 2.11.0. |
| HTTPX2 is a next generation HTTP client for Python. Prior to 2.11.0, Request._prepare() in src/httpx2/httpx2/_models.py can add a body-derived Content-Length header to a request that already contains a caller-supplied Transfer-Encoding header because its setdefault() processing checks each default header independently rather than treating the two framing headers as mutually exclusive. Fixed-size byte, JSON, form, and known-length multipart bodies can therefore be serialized over HTTP/1.1 with both headers, allowing request smuggling or connection desynchronization when downstream intermediaries disagree about which framing header takes precedence. This issue is fixed in version 2.11.0. |
| HTTPX2 is a next generation HTTP client for Python. Prior to 2.12.0, the HTTPX2 content decoders in src/httpx2/httpx2/_decoders.py fully inflate each gzip, deflate, br, or zstd network chunk before iter_bytes() or aiter_bytes() yields bounded pieces to the application. A 64 KiB compressed chunk can expand to approximately 64 MiB in one intermediate allocation, so an attacker-controlled or compromised server can cause severe memory pressure or out-of-memory process termination even when the application streams the response. This issue is fixed in version 2.12.0. |
| Pydantic AI is a Python agent framework for building Generative AI applications. In versions 1.56.0 through 1.98.0, when an application opts a URL into force_download='allow-local' (disabling the default block on private/internal IPs), the cloud-metadata blocklist could be bypassed by encoding the metadata IP in an IPv6 transition form (IPv4-mapped IPv6, 6to4, or NAT64), exposing cloud IAM short-term credentials on dual-stack or translated networks. This is an incomplete fix of GHSA-2jrp-274c-jhv3 / CVE-2026-25580, whose remediation did not hold for IPv6-encoded forms of the metadata IPs. An application is affected only if it explicitly opts a FileUrl (ImageUrl, AudioUrl, VideoUrl, DocumentUrl) into force_download='allow-local' on a URL influenced by untrusted input; it is not affected when using bundled integrations to ingest user input (Agent.to_web / clai web, VercelAIAdapter, AGUIAdapter / Agent.to_ag_ui), since they do not propagate force_download from external data, nor when downloading only from developer-controlled URLs. This issue has been fixed in version 1.99.0. |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. In versions 1.88.0 up to but not including 1.107.1 and 2.0.0b1 up to but not including 2.5.0, the UI adapters (AG-UI via Agent.to_ag_ui()/AGUIAdapter, and Vercel AI via VercelAIAdapter) use sanitize_messages to strip unresolved ("dangling") client-submitted tool calls from untrusted message history before it reaches the agent, a defense-in-depth default that prevents the agent from executing tool calls the model never emitted. However, the strip anchored to a message index computed before sanitization ran, so when a trailing client message sanitized to empty and was dropped (for example a client system message under the default manage_system_prompt='server'), a preceding assistant response carrying an unresolved tool call became the new tail and was dispatched without inspection. As a result, a remote client could cause a registered, non-approval server tool to run with client-supplied arguments rather than arguments the model produced. The impact is bounded by what the affected tools do and is most significant for applications that gate tool execution in a model-request hook (before_model_request / after_model_request), since a forged call skips the model turn and bypasses that guardrail; approval-gated tools (requires_approval=True) are not auto-executed by this path. This issue has been fixed in versions 1.107.1 and 2.5.0. |
| Pydantic AI is a Python agent framework for building Generative AI applications. In versions 1.65.0 through 1.105.0, and 2.0.0b1 through 2.0.0b5, a client that submits message history to a Pydantic AI UI adapter (such as the Vercel AI adapter) can reference arbitrary files in the application's model-provider or cloud-storage account. While file URL parts are validated against a scheme allowlist, UploadedFile references — which point to a file by provider file ID or cloud-storage URI (e.g. s3://…, gs://…) — were forwarded without validation. Because the provider resolves an UploadedFile using the server-side identity (IAM role, service account, or provider API key) rather than the client's, an attacker can craft message history to make the server read objects from its own account or other tenants, given a referenceable identifier. Exploitation requires a valid file identifier, which is not always unguessable depending on how the application names objects. This issue has been fixed in versions 1.106.0 and 2.0.0b6. |
| pydantic-settings provides settings management using Pydantic. From 2.12.0 until 2.14.2, NestedSecretsSettingsSource reads secret values from files in a configured secrets_dir. When secrets_nested_subdir=True, a directory entry inside secrets_dir that is a symbolic link pointing outside secrets_dir is followed, so files outside the configured directory are read into settings values. The same code path bypasses the documented secrets_dir_max_size protection. An attacker or lower-privileged component able to influence entries in the configured secrets directory (for example, a writable or shared secrets mount) can turn this into an unintended local file read into settings and can defeat the advertised loading-size cap. This vulnerability is fixed in 2.14.2. |
| Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. In versions 1.56.0 through 1.101.0, 2.0.0b1, and 2.0.0b2, the cloud-metadata blocklist could be bypassed by encoding the metadata IP in an IPv6 transition form that the previous fix, CVE-2026-46678, did not decode, exposing cloud IAM short-term credentials. The previous remediation decoded only IPv4-mapped IPv6, 6to4, and the NAT64 well-known prefix, so the metadata guarantee did not hold for the remaining transition forms: IPv4-compatible IPv6 (::a.b.c.d), the NAT64 RFC 8215 local-use prefix (64:ff9b:1::/48), operator-chosen NAT64 prefixes, and ISATAP. The IPv6 wrapper is then delivered to the underlying IPv4 metadata endpoint. This occurs when an application using Pydantic AI opts a URL into force_download='allow-local' (which disables the default block on private/internal IPs) and runs on a network that actually routes the affected IPv6 transition forms: NAT64-configured networks (IPv6-only or dual-stack-with-NAT64 deployments, including some Kubernetes setups) for the NAT64 variants, or networks with an ISATAP tunnel for ISATAP. A standard dual-stack cloud VM or container does not route these forms and is not affected in practice. The IPv4-compatible and Teredo variants are deprecated and addressed as defense-in-depth. This is an incomplete fix of GHSA-cqp8-fcvh-x7r3 / CVE-2026-46678 (itself a follow-up to CVE-2026-25580). This issue has been fixed in version 2.0.0b3. |
