<?xml version='1.0' encoding='UTF-8'?>
<?xml-stylesheet href="/static/style.xsl" type="text/xsl"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/" version="2.0">
  <channel>
    <title>Most recent entries from all</title>
    <link>https://vulnerability.circl.lu</link>
    <description>Contains only the most 10 recent entries.</description>
    <docs>http://www.rssboard.org/rss-specification</docs>
    <generator>python-feedgen</generator>
    <language>en</language>
    <lastBuildDate>Fri, 09 Oct 2026 20:32:35 +0000</lastBuildDate>
    <item>
      <title>BREW-oterm-CVE-2026-107294 — Pydantic AI: Unbounded memory use when downloading remote content via web_fetch or FileUrl</title>
      <link>https://vulnerability.circl.lu/vuln/brew-oterm-cve-2026-107294</link>
      <description>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; Homebrew: oterm&lt;/p&gt;
&lt;p&gt;### Summary&lt;/p&gt;
&lt;p&gt;Several remote-content download paths in Pydantic AI buffered the entire HTTP response body into memory before enforcing any size limit. An application that exposes the local web-fetch tool (`web_fetch_tool`, or the `WebFetch` capability&amp;#39;s local fallback) to untrusted prompts can be driven to fetch an attacker-chosen URL that streams a very large body, exhausting process memory and crashing the worker. The same unbounded buffering applied to `FileUrl` media downloads (`ImageUrl`, `DocumentUrl`, `VideoUrl`, `AudioUrl`).&lt;/p&gt;
&lt;p&gt;This is an **availability** issue only. SSRF protections (scheme allowlist, private-IP and cloud-metadata blocking) are unaffected; there is no confidentiality or integrity impact.&lt;/p&gt;
&lt;p&gt;### Details&lt;/p&gt;
&lt;p&gt;The download helpers read the full response body before applying content-size controls, so an existing text-length limit only truncated *after* the whole body was already in memory, and media downloads had no wire-level cap at all. A single large response could grow process memory without bound .&lt;/p&gt;
&lt;p&gt;### Who Is Affected&lt;/p&gt;
&lt;p&gt;You are affected if your application registers the local web-fetch tool (or relies on the `WebFetch` capability&amp;#39;s local fallback) and exposes the agent to untrusted prompts, or if it downloads large remote `FileUrl`s influenced by untrusted input. Applications that only fetch developer-controlled URLs are not exposed to the model-chosen attack path.&lt;/p&gt;
&lt;p&gt;### Remediation&lt;/p&gt;
&lt;p&gt;Upgrade to `2.24.0` or later (v2) or `1.107.2` or later (v1). Patched versi…&lt;/p&gt;</description>
      <content:encoded>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; Homebrew: oterm&lt;/p&gt;
&lt;p&gt;### Summary&lt;/p&gt;
&lt;p&gt;Several remote-content download paths in Pydantic AI buffered the entire HTTP response body into memory before enforcing any size limit. An application that exposes the local web-fetch tool (`web_fetch_tool`, or the `WebFetch` capability&amp;#39;s local fallback) to untrusted prompts can be driven to fetch an attacker-chosen URL that streams a very large body, exhausting process memory and crashing the worker. The same unbounded buffering applied to `FileUrl` media downloads (`ImageUrl`, `DocumentUrl`, `VideoUrl`, `AudioUrl`).&lt;/p&gt;
&lt;p&gt;This is an **availability** issue only. SSRF protections (scheme allowlist, private-IP and cloud-metadata blocking) are unaffected; there is no confidentiality or integrity impact.&lt;/p&gt;
&lt;p&gt;### Details&lt;/p&gt;
&lt;p&gt;The download helpers read the full response body before applying content-size controls, so an existing text-length limit only truncated *after* the whole body was already in memory, and media downloads had no wire-level cap at all. A single large response could grow process memory without bound .&lt;/p&gt;
&lt;p&gt;### Who Is Affected&lt;/p&gt;
&lt;p&gt;You are affected if your application registers the local web-fetch tool (or relies on the `WebFetch` capability&amp;#39;s local fallback) and exposes the agent to untrusted prompts, or if it downloads large remote `FileUrl`s influenced by untrusted input. Applications that only fetch developer-controlled URLs are not exposed to the model-chosen attack path.&lt;/p&gt;
&lt;p&gt;### Remediation&lt;/p&gt;
&lt;p&gt;Upgrade to `2.24.0` or later (v2) or `1.107.2` or later (v1). Patched versi…&lt;/p&gt;</content:encoded>
      <guid isPermaLink="false">https://vulnerability.circl.lu/vuln/brew-oterm-cve-2026-107294</guid>
    </item>
    <item>
      <title>fkie_cve-2026-107294</title>
      <link>https://vulnerability.circl.lu/vuln/fkie_cve-2026-107294</link>
      <description>&lt;p&gt;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.&lt;/p&gt;</description>
      <content:encoded>&lt;p&gt;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.&lt;/p&gt;</content:encoded>
      <guid isPermaLink="false">https://vulnerability.circl.lu/vuln/fkie_cve-2026-107294</guid>
    </item>
    <item>
      <title>GHSA-v2xh-2vp8-57h8 — Pydantic AI: Unbounded memory use when downloading remote content via web_fetch or FileUrl</title>
      <link>https://vulnerability.circl.lu/vuln/ghsa-v2xh-2vp8-57h8</link>
      <description>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; PyPI: pydantic-ai, PyPI: pydantic-ai-slim&lt;/p&gt;
&lt;p&gt;### Summary&lt;/p&gt;
&lt;p&gt;Several remote-content download paths in Pydantic AI buffered the entire HTTP response body into memory before enforcing any size limit. An application that exposes the local web-fetch tool (`web_fetch_tool`, or the `WebFetch` capability&amp;#39;s local fallback) to untrusted prompts can be driven to fetch an attacker-chosen URL that streams a very large body, exhausting process memory and crashing the worker. The same unbounded buffering applied to `FileUrl` media downloads (`ImageUrl`, `DocumentUrl`, `VideoUrl`, `AudioUrl`).&lt;/p&gt;
&lt;p&gt;This is an **availability** issue only. SSRF protections (scheme allowlist, private-IP and cloud-metadata blocking) are unaffected; there is no confidentiality or integrity impact.&lt;/p&gt;
&lt;p&gt;### Details&lt;/p&gt;
&lt;p&gt;The download helpers read the full response body before applying content-size controls, so an existing text-length limit only truncated *after* the whole body was already in memory, and media downloads had no wire-level cap at all. A single large response could grow process memory without bound .&lt;/p&gt;
&lt;p&gt;### Who Is Affected&lt;/p&gt;
&lt;p&gt;You are affected if your application registers the local web-fetch tool (or relies on the `WebFetch` capability&amp;#39;s local fallback) and exposes the agent to untrusted prompts, or if it downloads large remote `FileUrl`s influenced by untrusted input. Applications that only fetch developer-controlled URLs are not exposed to the model-chosen attack path.&lt;/p&gt;
&lt;p&gt;### Remediation&lt;/p&gt;
&lt;p&gt;Upgrade to `2.24.0` or later (v2) or `1.107.2` or later (v1). Patched versi…&lt;/p&gt;</description>
      <content:encoded>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; PyPI: pydantic-ai, PyPI: pydantic-ai-slim&lt;/p&gt;
&lt;p&gt;### Summary&lt;/p&gt;
&lt;p&gt;Several remote-content download paths in Pydantic AI buffered the entire HTTP response body into memory before enforcing any size limit. An application that exposes the local web-fetch tool (`web_fetch_tool`, or the `WebFetch` capability&amp;#39;s local fallback) to untrusted prompts can be driven to fetch an attacker-chosen URL that streams a very large body, exhausting process memory and crashing the worker. The same unbounded buffering applied to `FileUrl` media downloads (`ImageUrl`, `DocumentUrl`, `VideoUrl`, `AudioUrl`).&lt;/p&gt;
&lt;p&gt;This is an **availability** issue only. SSRF protections (scheme allowlist, private-IP and cloud-metadata blocking) are unaffected; there is no confidentiality or integrity impact.&lt;/p&gt;
&lt;p&gt;### Details&lt;/p&gt;
&lt;p&gt;The download helpers read the full response body before applying content-size controls, so an existing text-length limit only truncated *after* the whole body was already in memory, and media downloads had no wire-level cap at all. A single large response could grow process memory without bound .&lt;/p&gt;
&lt;p&gt;### Who Is Affected&lt;/p&gt;
&lt;p&gt;You are affected if your application registers the local web-fetch tool (or relies on the `WebFetch` capability&amp;#39;s local fallback) and exposes the agent to untrusted prompts, or if it downloads large remote `FileUrl`s influenced by untrusted input. Applications that only fetch developer-controlled URLs are not exposed to the model-chosen attack path.&lt;/p&gt;
&lt;p&gt;### Remediation&lt;/p&gt;
&lt;p&gt;Upgrade to `2.24.0` or later (v2) or `1.107.2` or later (v1). Patched versi…&lt;/p&gt;</content:encoded>
      <guid isPermaLink="false">https://vulnerability.circl.lu/vuln/ghsa-v2xh-2vp8-57h8</guid>
    </item>
  </channel>
</rss>
