Reference page — cumulative record through Monday, October 19, 2026 UTC. Reference pages update as the archive grows; only dated daily editions are immutable pages of record.
CVE-2026-107289
Pydantic AI: SSRF cloud-metadata blocklist bypass via IPv6 zone identifier (incomplete fix for CVE-2026-46678 and CVE-2026-48782)
AV AC PR UI S C I A CVSS EPSS %ile KEV
N H N N C H N N 6.8 .0032 23.7 —
AFFECTED
Product Versions Fixed
pydantic-ai >= 1.56.0, < 1.107.6 – —
pydantic-ai-slim >= 1.56.0, < 1.107.6 – —
TIMELINE
Oct 7 Reserved by GitHub_M
Oct 8 Published (CNA: GitHub_M)
Oct 13 EXPLOIT PUBLISHED — CVE-2026-107289 (pydantic-ai). Public exploit reference added.
Description
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.
Lifecycle
Complete event history — 3 events, chronological
| Date | Event | Detail |
|---|
| October 7, 2026 | Reserved | Reserved by GitHub_M |
| October 8, 2026 | Published | Published (CNA: GitHub_M) |
| October 13, 2026 | EXPLOIT PUBLISHED | EXPLOIT PUBLISHED — CVE-2026-107289 (pydantic-ai). Public exploit reference added. |
Affected
Affected products and packages — 2 rows
| Vendor | Product / Package | Ecosystem | Version introduced | Fixed |
|---|
| pydantic | pydantic-ai | — | >= 1.56.0, < 1.107.6 | — |
| pydantic | pydantic-ai-slim | — | >= 1.56.0, < 1.107.6 | — |
References (7)
- https://github.com/pydantic/pydantic-ai/commit/02157e1b87bd45d3f2e111ce07afdf89f9fb0e5b [Exploit, Patch]
- https://github.com/pydantic/pydantic-ai/commit/4da70591460f51a8c4f128eaeef70a33340dbd55 [Exploit, Patch]
- https://github.com/pydantic/pydantic-ai/pull/8401 [Exploit, Patch, Vendor Advisory]
- https://github.com/pydantic/pydantic-ai/pull/8402 [Patch, Vendor Advisory]
- https://github.com/pydantic/pydantic-ai/releases/tag/v1.107.6 [Patch, Release Notes, Vendor Advisory]
- https://github.com/pydantic/pydantic-ai/releases/tag/v2.44.0 [Patch, Release Notes, Vendor Advisory]
- https://github.com/pydantic/pydantic-ai/security/advisories/GHSA-vmxc-h2x2-jmf3 [Exploit, Mitigation, Vendor Advisory]
About this page
This is a reference page, not a dated page of record. It assembles the complete lifecycle of CVE-2026-107289 from the CVE Program record, NVD enrichment, the CISA KEV catalog, EPSS, and OSV advisories. The box score's numbers (CVSS, EPSS, KEV status) are current as of Monday, October 19, 2026 UTC and are re-derived as the archive grows; only dated daily editions are immutable pages of record. The authoritative source for this identifier is cve.org.
