# CVE-2026-107289: Pydantic AI: SSRF cloud-metadata blocklist bypass via IPv6 zone identifiers > Attackers can bypass cloud metadata blocklists in Pydantic AI via IPv6 zone identifiers to steal IAM credentials when local URL fetching is enabled. - **Severity:** Medium 6.8 (`CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:N/A:N`) - **Exploited in the wild:** Not known - **Login needed:** Not needed - **Attackable:** Over the internet - **Victim action needed:** Not needed - **Fix available:** Yes - **Published:** 8 Oct 2026 - **Updated:** 8 Oct 2026 - **Also known as:** CLEANSTART-2026-OZ89271, GHSA-vmxc-h2x2-jmf3, PYSEC-2026-4282, PYSEC-2026-4291 ## What is CVE-2026-107289? CVE-2026-107289 is a medium-severity server-side request forgery vulnerability in pydantic-ai, published on 8 Oct 2026. An attacker can exploit it over the internet without logging in, with no action from the victim. ### Official 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. Weakness: CWE-918 (Server-Side Request Forgery), CWE-1289 (Improper Validation of Unsafe Equivalence in Input) ## Which versions are affected by CVE-2026-107289? 10 open source packages ship an affected version, including pydantic-ai and pydantic-ai-slim. It affects 1 product. | Package | Ecosystem | Affected | Fixed in | | --- | --- | --- | --- | | `pydantic-ai` | PyPI | From 1.56.0 | 1.107.6, 2.44.0 | | `pydantic-ai` | PyPI | From 1.56.0 | 1.107.6 | | `pydantic-ai` | PyPI | From 1.56.0 | 1.107.6 | | `pydantic-ai` | PyPI | From 2.0.0b1 | 2.44.0 | | `pydantic-ai` | PyPI | From 2.0.0b1 | 2.44.0 | | `pydantic-ai-slim` | PyPI | From 1.56.0 | 1.107.6, 2.44.0 | | `pydantic-ai-slim` | PyPI | From 1.56.0 | 1.107.6 | | `pydantic-ai-slim` | PyPI | From 1.56.0 | 1.107.6 | | `pydantic-ai-slim` | PyPI | From 2.0.0b1 | 2.44.0 | | `pydantic-ai-slim` | PyPI | From 2.0.0b1 | 2.44.0 | Products: Pydantic Pydantic Ai. ## How do I fix CVE-2026-107289? Upgrade pydantic-ai to a fixed version for your release line: 1.107.6 or 2.44.0. ## Is CVE-2026-107289 being exploited? There is no known exploitation of CVE-2026-107289 in the wild and no public exploit code. ## Other vulnerabilities in pydantic-ai - [CVE-2026-25580](https://insights.sbomify.com/CVE-2026-25580/): Pydantic AI Affected by Server-Side Request Forgery (SSRF) in URL Download Handling (high 8.6) - [CVE-2026-107295](https://insights.sbomify.com/CVE-2026-107295/): Pydantic AI Web chat UI (`Agent.to_web()`, `clai web`): a website visited by the developer can trigger agent runs and server-side tool execution on the local chat endpoint (high 7.6) - [CVE-2026-107286](https://insights.sbomify.com/CVE-2026-107286/): Pydantic AI: Concurrency-limited models can keep their slot when a streamed request ends early (high 7.5) - [CVE-2026-25640](https://insights.sbomify.com/CVE-2026-25640/): Pydantic AI affected by Stored XSS via Path Traversal in Web UI CDN URL (high 7.1) - [CVE-2026-46678](https://insights.sbomify.com/CVE-2026-46678/): Pydantic AI: SSRF cloud-metadata blocklist bypass via IPv4-mapped IPv6 (Incomplete fix of CVE-2026-25580) (medium 6.8) ## Fix prompt for AI coding agents ```text Check this codebase for CVE-2026-107289 and fix it if it is affected. What it is: CVE-2026-107289 is a medium-severity server-side request forgery vulnerability in pydantic-ai, published on 8 Oct 2026. An attacker can exploit it over the internet without logging in, with no action from the victim. Exploitation: There is no known exploitation of CVE-2026-107289 in the wild and no public exploit code. Affected packages and fixed versions: - pydantic-ai (PyPI): affected from 1.56.0; fixed in 1.107.6, 2.44.0 - pydantic-ai (PyPI): affected from 1.56.0; fixed in 1.107.6 - pydantic-ai (PyPI): affected from 1.56.0; fixed in 1.107.6 - pydantic-ai (PyPI): affected from 2.0.0b1; fixed in 2.44.0 - pydantic-ai (PyPI): affected from 2.0.0b1; fixed in 2.44.0 - pydantic-ai-slim (PyPI): affected from 1.56.0; fixed in 1.107.6, 2.44.0 - pydantic-ai-slim (PyPI): affected from 1.56.0; fixed in 1.107.6 - pydantic-ai-slim (PyPI): affected from 1.56.0; fixed in 1.107.6 - pydantic-ai-slim (PyPI): affected from 2.0.0b1; fixed in 2.44.0 - pydantic-ai-slim (PyPI): affected from 2.0.0b1; fixed in 2.44.0 Task: 1. Search the dependency manifests and lockfiles (for example package.json, package-lock.json, requirements.txt, poetry.lock, uv.lock, pom.xml, build.gradle, go.mod, Cargo.lock, Gemfile.lock) and any Dockerfiles for the affected packages, including transitive dependencies. 2. If an affected version is in use, upgrade it to the fixed version for its release line with the project's package manager, so the lockfile is regenerated. Prefer the smallest upgrade that fixes it. 3. If there is no fix, or the upgrade would break the build, do not guess: explain the risk and the safest workaround instead. 4. Run the tests and summarise what changed and why. 5. If nothing in the project is affected, say so and change nothing. Reference: https://insights.sbomify.com/CVE-2026-107289/ (Markdown: https://insights.sbomify.com/CVE-2026-107289/index.md) ``` ## Sources - CVE record: https://www.cve.org/CVERecord?id=CVE-2026-107289 - NVD: https://nvd.nist.gov/vuln/detail/CVE-2026-107289 - OSV: https://osv.dev/vulnerability/CVE-2026-107289 --- From sbomify Insights: https://insights.sbomify.com/CVE-2026-107289/ Summaries are generated and may contain mistakes; the official description above is the record.