CVE-2026-48782Medium· 6.8▾ Sunlitpydantic-ai: SSRF blocklist bypass via IPv4-compatible, SIIT/IVI, and local NAT64 IPv6 addresses (incomplete fix of CVE-2026-46678)
▾ Sunlit zone — Low / medium · no exploitation signal
impact 37.4 · likelihood 0.1 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
Exploit-prediction probability, daily snapshots since Jul 4.
Disclosure to exploitation, from the record and what we observed since indexing it.
Disclosed via GHSA
0.3%
0.3% → 0.3%
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 routes the affected IPv6 transition forms (NAT64- or ISATAP-configured networks), the cloud-metadata blocklist could be bypassed by encoding the metadata IP in an IPv6 transition form that the previous fix did not decode — IPv4-compatible IPv6 (::a.b.c.d), the NAT64 RFC 8215 local-use prefix (64:ff9b:1::/48), operator-chosen NAT64 prefixes, or ISATAP. The IPv6 wrapper is then delivered to the underlying IPv4 metadata endpoint, exposing cloud IAM short-term credentials.
The bypass is exploitable only in environments whose network actually routes these 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 them 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). The prior remediation decoded only IPv4-mapped IPv6, 6to4, and the NAT64 well-known prefix; the metadata guarantee did not hold for the remaining transition forms.
MEDIUM — CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:N/A:N = 6.8
Same impact metrics and narrow attack surface as the parent advisory (AC:H): exploitation requires the application to have opted into allow-local on a URL influenced by untrusted input, and the NAT64/ISATAP variants additionally require the deployment network to route those forms.
CWE-918: Server-Side Request Forgery (SSRF)
| Package | Vulnerable | Patched |
|---|---|---|
pydantic-ai | >= 1.56.0, < 1.102.0; >= 2.0.0b1, < 2.0.0b3 | 1.102.0; 2.0.0b3 |
pydantic-ai-slim | >= 1.56.0, < 1.102.0; >= 2.0.0b1, < 2.0.0b3 | 1.102.0; 2.0.0b3 |
These transition forms have not been decoded since SSRF protection was introduced in 1.56.0.
Users are affected only if their application explicitly opts a FileUrl (ImageUrl, AudioUrl, VideoUrl, DocumentUrl) into force_download='allow-local' on a URL that is, or could be, influenced by untrusted input.
Beyond that precondition, the affected encodings only reach a metadata endpoint in environments whose network actually routes them. The broadly-routable IPv4-mapped form was addressed in 1.99.0 (CVE-2026-46678); the additional forms addressed here require a NAT64-configured network (IPv6-only or dual-stack-with-NAT64 deployments, including some Kubernetes setups) for the NAT64 variants, or an ISATAP tunnel for the ISATAP variant. The IPv4-compatible and Teredo forms are deprecated and not routed by modern stacks; they are addressed as defense-in-depth. Most deployments on a standard dual-stack cloud VM or container are therefore not exploitable in practice, but the fix restores the "always blocked" guarantee for the environments that are.
Users are not affected if they use any of the bundled integrations to ingest user input, because they do not propagate force_download from external data:
Agent.to_web / clai webVercelAIAdapterAGUIAdapter / Agent.to_ag_uiApplications that only download from developer-controlled URLs are not affected.
Upgrade to 1.102.0 or later (or 2.0.0b3 or later on the 2.0 pre-release line). The cloud-metadata and private-IP blocklists now decode the embedded IPv4 of every standardized IPv6 transition form before evaluating it — IPv4-mapped, IPv4-compatible, 6to4, NAT64 across all prefix lengths (including the RFC 8215 local-use prefix and operator-chosen prefixes), ISATAP, and Teredo. The set of always-blocked cloud metadata/credential endpoints has also been expanded across providers.
Avoid passing force_download='allow-local' on any URL that could be influenced by untrusted input. If developers must, resolve the hostname themselves and validate the result against their own metadata blocklist — including IPv6 transition forms — before constructing the FileUrl.
Reported by @SnailSploit.
pydantic-ai-slim >= 1.56.0, < 1.102.0pydantic-ai >= 1.56.0, < 1.102.0pydantic-ai >= 2.0.0b1, < 2.0.0b3pydantic-ai-slim >= 2.0.0b1, < 2.0.0b3Upgrade to a patched release:
pydantic-ai-slim 1.102.0pydantic-ai 1.102.0pydantic-ai 2.0.0b3pydantic-ai-slim 2.0.0b3Connected by shared product, vendor, weakness, or advisory.
CVE-2026-54249Medium· 6.8Pydantic AI: Unvalidated UploadedFile references in UI adapters allow server-side file access using the application's credentials
CVE-2025-68616High· 7.5WeasyPrint helps web developers to create PDF documents
CVE-2026-12992High· 7.4A flaw was found in Apicurio Registry
CVE-2021-21985Critical· 9.8The vSphere Client (HTML5) contains a remote code execution vulnerability due to lack of input validation in the Virtual SAN Health Check plug-in which is enabled by default in vCenter Server
CVE-2021-21975High· 7.5Server Side Request Forgery in vRealize Operations Manager API (CVE-2021-21975) prior to 8.4 may allow a malicious actor with network access to the vRealize Operations Manager API can perform a Server Side Request Forgery attack to steal…
CVE-2021-26855Critical· 9.1Microsoft Exchange Server Remote Code Execution Vulnerability